Michael Blauth, Stephen L Kates, Joseph A Nicholas
Osteoporotic Fracture Care Medical and Surgical Management
Table of Contents Foreword
V
Section 1—Principles
1
1.1 P rinciples of orthogeriatric medical care Preface
VI
Joseph A Nicholas
3
1.2 P rinciples of orthogeriatric surgical care Acknowledgments
VII
Michael Blauth
7
1.3 P rinciples of orthogeriatric anesthesia Contributors
VIII
Abbreviations
XII
Online AO Educational Content
XV
Ali Shariat, Malikah Latmore
19
1.4 P reoperative risk assessment and preparation Joseph A Nicholas
29
1.5 P rognosis and goals of care Joshua Uy
35
1.6 A nticoagulation in the perioperative setting Table of contents
XVI
Lauren J Gleason, Adeela Cheema, Joseph A Nicholas
41
1.7 Postoperative medical management Jennifer D Muniak, Susan M Friedman
51
1.8 Postoperative surgical management Section 1—Principles
1
Michael Blauth, Peter Brink
59
1.9 Postacute care Section 2—Improving the system of care
115
Section 3—Fracture management
189
Bernardo Reyes, Nemer Dabage, Darby Sider
71
1.10 O steoporosis Rashmi Khadilkar, Krupa Shah
77
1.11 S arcopenia, malnutrition, frailty, and falls Claudia M Gonzalez Suarez
89
1.12 Pain management Index
593
Timothy Holahan, Daniel A Mendelson
97
1.13 Polypharmacy Bernardo Reyes, Justinder Malhotra
105
1.14 D elirium Markus Gosch, Katrin Singler
XVI
109
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Section 2—Improving the system of care
115
2.1 M odels of orthogeriatric care Andrea Giusti, Giulio Pioli
Clemens Hengg, Vajara Phiphobmongkol 129
2.6 O rthogeriatric team—principles, roles, and responsibilities Markus Gosch, Michael Blauth
2.7 P rotocol and order set development Stephen L Kates, Joseph A Nicholas
157
2.8 F racture liaison service and improving treatment rates for osteoporosis Paul J Mitchell
165
315
3.7 Pelvic ring Pol M Rommens, Michael Blauth, Alexander Hofmann
339
3.8 A cetabulum Dietmar Krappinger, Richard A Lindtner, Herbert Resch
373
Simon C Mears, Stephen L Kates
389
3.10 Trochanteric and subtrochanteric femur 173
Carl Neuerburg, Christian Kammerlander, Stephen L Kates
405
3.11 F emoral shaft
2.10 L ean business principles Stephen L Kates, Andrew J Pugely
297
3.9 F emoral neck
2.9 U se of registry data to improve care Colin Currie
283
3.6 D istal forearm Rohit Arora, Alexander Keiler, Susanne Strasser
151
269
3.5 O lecranon Peter Kaiser, Simon Euler
145
243
3.4 Elbow Rohit Arora, Kerstin Simon, Marco Keller, Michael Blauth
137
2.5 A dapting facilities to fragility fracture patients Edgar Mayr
191
3.3 D istal humerus Rohit Arora, Alexander Keiler, Michael Blauth
133
2.4 Elements of an orthogeriatric comanaged program Carl Neuerburg, Christian Kammerlander
Franz Kralinger, Michael Blauth
3.2 H umeral shaft
2.3 C linical practice guidelines Stephen L Kates, Michael Blauth
189
3.1 P roximal humerus 117
2.2 O vercoming barriers to implementation of a care model Stephen L Kates
Section 3—Fracture m anagement
181
Elizabeth B Gausden, Dean G Lorich
421
3.12 D istal femur Jong-Keon Oh, Christoph Sommer
439
3.13 Periprosthetic fractures around the hip Steven Velkes, Karl Stoffel
461
3.14 Periprosthetic fractures around the knee Frank A Liporace, Iain McFadyen, Richard S Yoon
479
3.15 P roximal tibia Michael Götzen, Michael Blauth
501
3.16 T ibial shaft Björn-Christian Link, Philippe Posso, Reto Babst
523
3.17 A nkle Christian CMA Donken, Michael HJ Verhofstad
535
3.18 A typical fractures Chang-Wug Oh, Joon-Woo Kim
559
3.19 C hest trauma Hans-Christian Jeske
571
3.20 Polytrauma Julie A Switzer, Herman Johal
579
XVII
Joseph A Nicholas
1.1 Principles of orthogeriatric medical care Joseph A Nicholas
1
Introduction
Despite the large amount of surgical care delivered to older adults [1], perioperative practice remains inappropriately anchored to the surgical experience of more robust and less comorbid patients. At best, many common and accepted approaches to specific illnesses are ineffective in older adults, and at worst, these practices contribute to serious morbidity and mortality [2, 3]. The negative impact of usual medical and surgical care is most pronounced in frail and medically complicated patients [4, 5]. The typical fragility fracture patient (FFP) is emblematic of patients for whom usual medical care is often the wrong care. To those who treat and research this population, it is not surprising that superior postoperative outcomes have been found through unique clinical and systems approaches to the geriatric patient [6, 7], strategies that often diverge from the types of medical investigations and treatments used in most care settings. Fortunately, there is growing evidence that improved clinical outcomes can be obtained in frail older adults with osteoporotic fractures through the incorporation of a relatively small number of standard approaches and clinical pathways [8]. The major barriers to implementing these approaches are not technological or financial but involve an understanding and commitment to creating systems and expertise that focus on standardizing care, avoiding adverse events, and adapting treatments to the unique physiology and prognosis of the older adult. While the details of such care will change as the evidence base expands, we expect the basic strategies outlined in this book to remain relevant for years to come. In the chapters that follow, readers will be introduced to the principles and specifics of caring for the typical FFP, based on the improved outcomes produced by orthogeriatric comanagement in organized fracture center programs. To set the stage, there are a number of principles that are important to recognize.
2
Key principles
2.1
lder adults are not simply adults with more O illnesses
Compared with younger adults, older adults have unique physiologies, regardless of the presence or absence of specific comorbidities [9, 10]. Aging results in biological changes that render the older adult more susceptible to the harms of immobility, diagnostic tests, and medication effects. For this reason, many common medical practices can be ineffective or harmful in older adults. Examples include exaggerated hypotension in the presence of anesthetics and blood loss, low thresholds for delirium, complications due to polypharmacy, and rapid functional decline with immobility. This general decreased ability to respond to physiological stress is best described as frailty [11]. 2.2
ip fracture surgery can be performed safely and H effectively even on frail patients
High-performing hip fracture centers produce low shortterm mortality rates (ie, less than 2%), even in populations with high degrees of frailty and comorbidity [6, 12]. Advances in anesthesia, implant technology allowing for e arly weight bearing as tolerated, orthopedic procedural improvements, and orthogeriatric comanagement all contribute to rapid, safe, and effective repair of the overwhelming majority of hip fracture patients. Urgent surgery in the optimized patient is now standard care to avoid the short-term harms of ongoing pain, blood loss, and immobility. 2.3
ge is not the most important indicator of risk or A prognosis in hip fracture patients
While age is a general predictor for outcomes and complications, it is more helpful to base risk assessments and treatment decisions on functional status, cognitive status, and comorbidity [13]. Asking patients about their day-to-day life can help estimate operative risk, recovery potential, and life expectancy better than disease-based assessments.
3
Section 1 Principles 1.1 Principles of orthogeriatric medical care
2.4
urgical delay and immobility leads to S irreversible muscle loss in the older adult
Early surgery is superior [14] and essential for frail and comorbid patients. The medical and surgical team must constantly weigh the impact of functional decline and operative delay against operative risk. Even the frailest patients can usually be optimized quickly, repaired, and begin immediate full weight bearing and rehabilitation [15]. 2.5
G et the patient moving as soon as possible
Because rapid loss of muscle mass and function is a fundamental issue resulting in poor overall outcomes [16], all care pathways should be optimized to support early mobility and rehabilitation. While surgical delay and bed rest orders are obvious factors, polypharmacy, excessive testing, frequent subspecialty consultation, and inadequate pain control are all common barriers to mobilization that need to be minimized. Early mobility provides the necessary physical and emotional stimulation [17] for healing and recovery and helps minimize skin breakdown, constipation, and neuromuscular wasting. Mobility can be the difference between rapid recovery and prolonged hospitalization. 2.6
any surgeons, internists, and specialists do not M understand typical geriatric medical physiology
Regardless of professional training, unique geriatric responses to therapies are not adequately emphasized in most medical school and postgraduate training programs [19, 20]. Clinical experiences in geriatrics often fail to focus on acute care approaches, and subspecialty training in many medical and surgical disciplines does not typically promote adaptation of clinical expertise to frail older adults [21]. Competency in acute geriatric care does not require formal fellowship training, but can be achieved with a continuing medical education approach. Attending a course, viewing educational media, or visiting an established geriatric fracture program can help develop competency in caring for older adults.
4
any geriatricians, internists, and specialists do M not understand acute perioperative medicine
Current medical training offers little focus on the perioperative period. Other than performing outpatient preoperative risk assessments in relatively robust patients or planning an elective procedure, most internists, subspecialists, and geriatricians do not gain expertise in acute stabilization, optimization, and recovery of patients undergoing urgent surgery. Approaches to common medical issues are different in perioperative patients from those in typical medical admissions [22]. 2.9
ery little high-quality evidence is applicable to V the care of older adults
Most medical and surgical evidence is based on adults that are very different from the geriatric fracture patient [23]. Geriatric populations do not experience the same balance of benefits and harms younger, healthier, and more robust adults do. Rather than trying to comply with multiple diseasespecific guidelines, high-performing geriatric fracture centers create strategies based on general geriatric principles, like avoiding polypharmacy, anticipating and managing delirium, and rapid restoration of mobility.
Less is often more
Most FFPs have multiple comorbidities and abnormalities on diagnostic testing, many of which are chronic, clinically irrelevant, or unable to be improved. Unfortunately, this often results in excessive testing and consultation, overdiagnosis, and polypharmacy. Organized programs work hard to avoid these distractions, and focus instead on key areas like hemodynamic stability, pain control, prompt fracture reduction, and mobilization [18]. 2.7
2.8
2.10 Recognize failing patients at the end of life
For many patients, falls and fragility fractures are the result of decompensated medical illnesses and frailty, and many will have a life expectancy of less than 6 months [24]. Failing patients do not respond well to usual medical care, suffering more harm than benefits from hospitalization, testing, and treatment. Early recognition of failing patients is important to identify achievable goals, set realistic expectations for the family and the clinical team, and to focus future care appropriately on end of life. Orthopedic surgery plays an essential role in pain control and quality of life. All clinicians involved in the care of FFPs need to have an ability to recognize the failing patient (ie, frailty). 2.11
Organized fracture programs work
There is no single surgical technique, preoperative risk assessment tool, or standard medical consultation that will produce ongoing results as good as an organized approach to the FFP. Investments in an organized program with geriatric comanagement will yield improvement in outcomes, costs, and both patient and physician satisfaction [8, 25]. Organized programs are becoming the standard of care in many medical and surgical communities [26], and even for other surgical problems [27, 28].
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Joseph A Nicholas
3
References
1. Centers for Disease Control and Prevention. Number of discharges from short-stay hospitals, by first-listed diagnosis and age: United States, 2010. Available at: www.cdc.gov/nchs/data/ nhds/3firstlisted/2010first3_ numberage.pdf. Accessed April 15, 2017. 2. Gerstein HC, Miller ME, Byington RP, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008 Jun 12;358(24):2545–2559. 3. Mossello E, Pieraccioli M, Nesti N, et al. Effects of low blood pressure in cognitively impaired elderly patients treated with antihypertensive drugs. JAMA Intern Med. 2015 Apr;175(4):578–585. 4. Odden MC, Peralta CA, Haan MN, et al. Rethinking the association of high blood pressure with mortality in elderly adults: the impact of frailty. Arch Intern Med. 2012 Aug 13;172(15):1162–1168. 5. Mosquera C, Spaniolas K, Fitzgerald TL. Impact of frailty on surgical outcomes: The right patient for the right procedure. Surgery. 2016 Aug;160(2):272–280. 6. Friedman SM, Mendelson DA, Bingham KW, et al. Impact of a comanaged Geriatric Fracture Center on short-term hip fracture outcomes. Arch Intern Med. 2009 Oct 12;169(18):1712–1717. 7. Kammerlander C, Gosch M, Blauth M, et al. The Tyrolean Geriatric Fracture Center: an orthogeriatric comanagement model. Z Gerontol Geriatr. 2011 Dec;44(6):363–367. 8. Friedman SM, Mendelson DA, Kates SL, et al. Geriatric co-management of proximal femur fractures: total quality management and protocol-driven care result in better outcomes for a frail patient population. J Am Geriatr Soc. 2008 Jul;56(7):1349–1356. 9. Cheitlin MD. Cardiovascular physiology-changes with aging. Am J Geriatr Cardiol. 2003 Jan–Feb;12(1):9–13.
10. Walston J, Hadley EC, Ferrucci L, et al. Research agenda for frailty in older adults: toward a better understanding of physiology and etiology: summary from the American Geriatrics Society/ National Institute on Aging Research Conference on Frailty in Older Adults. J Am Geriatr Soc. 2006 Jun;54(6):991–1001. 11. Fried LP, Tangen CM, Walston J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001 Mar;56(3):M146–M156. 12. Brauer CA, Coca-Perraillon M, Cutler DM, et al. Incidence and mortality of hip fractures in the United States. JAMA. 2009 Oct 14;302(14):1573–1579. 13. Semel J, Gray JM, Ahn HJ, et al. Predictors of outcome following hip fracture rehabilitation. PM R. 2010 Sep;2(9):799–805. 14. Lefaivre KA, Macadam SA, Davidson DJ, et al. Length of stay, mortality, morbidity and delay to surgery in hip fractures. J Bone Joint Surg Br. 2009 Jul;91(7):922–927. 15. Keehan R, Rees D, Kendrick E, et al. Enhanced recovery for fractured neck of femur: a report of 3 cases. Geriatr Orthop Surg Rehabil. 2014 Jun;5(2):37–42. 16. Visser M, Harris TB, Fox KM, et al. Change in muscle mass and muscle strength after a hip fracture: relationship to mobility recovery. J Gerontol A Biol Sci Med Sci. 2000 Aug;55(8):M434–M440. 17. Kalisch BJ, Lee S, Dabney BW. Outcomes of inpatient mobilization: a literature review. J Clin Nurs. 2014 Jun;23(11–12):1486–1501. 18. Nicholas JA. Preoperative optimization and risk assessment. Clin Geriatr Med. 2014 May;30(2):207–218. 19. Monette M, Hill A. Arm-twisting medical schools for core geriatric training. CMAJ. 2012 Jul 10;184(10):E515–E516.
20. Sedhom R, Barile D. Teaching our doctors to care for the elderly: a geriatrics needs assessment targeting internal medicine residents. Gerontol Geriatr Med. 2017 Jan–Dec;3:2333721417701687. 21. Potter JF, Burton JR, Drach GW, et al. Geriatrics for residents in the surgical and medical specialties: implementation of curricula and training experiences. J Am Geriatr Soc. 2005 Mar;53(3):511–515. 22. Braude P, Partridge JS, Hardwick J, et al. Geriatricians in perioperative medicine: developing subspecialty training. Br J Anaesth. 2016 Jan;116(1):4–6. 23. Tinetti ME, Bogardus ST Jr, Agostini JV. Potential pitfalls of disease-specific guidelines for patients with multiple conditions. N Engl J Med. 2004 Dec 30;351(27):2870–2874. 24. Neuman MD, Silber JH, Magaziner JS, et al. Survival and functional outcomes after hip fracture among nursing home residents. JAMA Intern Med. 2014 Aug;174(8):1273–1280. 25. Kates SL, Mendelson DA, Friedman SM. The value of an organized fracture program for the elderly: early results. J Orthop Trauma. 2011 Apr;25(4):233–237. 26. Fisher AA, Davis MW, Rubenach SE, et al. Outcomes for older patients with hip fractures: the impact of orthopedic and geriatric medicine cocare. J Orthop Trauma. 2006 Mar;20(3):172–178; discussion 179–180. 27. Tadros RO, Faries PL, Malik R, et al. The effect of a hospitalist comanagement service on vascular surgery inpatients. J Vasc Surg. 2015 Jun;61(6):1550–1555. 28. Montero Ruiz E, Rebollar Merino A, Rivera Rodriguez T, et al. Effect of comanagement with internal medicine on hospital stay of patients admitted to the Service of Otolaryngology. Acta Otorrinolaringol Esp. 2015 Sep–Oct;66(5):264–268.
5
Section 1 Principles 1.1 Principles of orthogeriatric medical care
6
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Michael Blauth
1.2 Principles of orthogeriatric surgical care Michael Blauth
1
Introduction
Fragility fracture patients (FFPs) represent up to 40% of patients in many orthopedic trauma units worldwide. This trend is increasing. As a consequence, over the last decade, refined surgical care approaches have been developed from growing experience and close collaboration with geriatricians in order to improve patient outcomes and lower healthcare expenses. Similar to fracture care in children, geriatric fracture care also differs in many aspects from the standard treatment of middle-aged adults. Due to the relative paucity of randomized trial data for many treatments, many of the following recommendations represent expert opinions with some based on biomechanical or clinical investigations. The four AO Principles certainly apply to the care of fragility fractures and should be carefully adhered to: 1. Fracture reduction and fixation to restore anatomical relationships 2. Stability by fixation or splinting, as the personality of the fracture and the injury requires 3. Preservation of blood supply to soft tissues and bone by careful handling and gentle reduction techniques 4. Early and safe mobilization of the part and the patient
2
Goal setting
The entire patient must be considered including his/her medical problems, medications, living situation, and goals for care. Overall, the following issues assume prominence in care of FFPs: • Pain relief • Prevention of functional decline • Maintenance of independence
• Prevention of complications, such as reoperations, pneumonia, pressure sores, urinary tract infection, and delirium Making the right therapeutic decisions is much more complex than with younger patients. Fragility fracture patients are functionally and physiologically variable (from non ambulatory “No-goes” to ambulatory “Go-goes”) that the benefits and risks of treatment are not as clear as in younger patients. Therefore, it is essential to establish a consensus for the treatment goals among all of the team members. Defining individual goals for each FFP is an important step which should be established and agreed upon as early as possible by the interdisciplinary team. The individual goals influence diagnostic and therapeutic surgical and medical measures and should be clearly communicated. Goal setting avoids unnecessary steps and streamlines the treatment. Goals may be adjusted during the treatment process. First, treatment goals should be very specific, clear and easy. Second, if you cannot measure it, you cannot manage it. Third, a goal needs to be attractive and acceptable to the patient and the clinical team. Fourth, the goal should be realistic, meaning achievable or “doable”. Fifth, the timeline to achieve the goal should be considered by setting a time frame. It is useful to find short-term as well as long-term goals. Usually, the long-term goal is the expected outcome in several weeks or months, like to live independently or to walk without using a walking aid. When approaching a long-term goal, you need different short-term goals for each problem, like walking with a rolling walker after the first week, or removing a urinary catheter within 2 or 3 days after surgery. The goals may be modified due to medical or surgical complications or if patients become unwilling or unable to continue or if they progress more slowly or quickly than expected. Goal setting should be integrated in the regular team meetings.
7
Section 1 Principles 1.2 Principles of orthogeriatric surgical care
3
Time matters
The decision-making process regarding the definitive treatment in complex situations or relative indications is often delayed for multiple reasons. Goal setting and standardized communication pathways help to avoid unnecessary delay and expedite treatment.
Most studies suggest that performing surgery within the first 24–48 hours of admission decreases the number of complications and mortality. Delays longer than 72 hours are associated with an increased risk of multiple complications and mortality.
4
Surgical fixation reduces pain and blood loss significantly. It is also unethical to unnecessarily delay surgery.
The musculoskeletal system of older patients is more vulnerable to problems and less tolerant of stress:
The earlier surgical stabilization is performed, the better. This guiding principle is often violated because of the patient condition, patient consent, or hospital system barriers. The system of care must be optimized to avoid delay and iatrogenic problems.
• Skin may be thin and less elastic due to atrophy or malnutrition and making pressure sores and degloving injuries more common. Wounds in older adults may also heal poorly for similar reasons. During positioning and draping, the surgeon must remember that the older patient’s skin is fragile and can tear or be avulsed with minimal shear stresses. Shear forces from manual traction, removal of surgical drapes or localized pressure by splints and traction devices must be avoided (Fig 1.2-1). In surgery, meticulous positioning helps avoid skin breakdown.
The operating time should be as short as possible to reduce the stresses of surgery and its burdens on the patient.
a
h
b
c
Soft-tissue conditions
d
e
f
g
i
Fig 1.2-1a–i a–c An 88-year-old woman with a type B2 periprosthetic femoral fracture. d–g Revision hemiarthroplasty (d), follow-up at 2 months (e–g). h After removing the covers, a degloving of the lower leg skin by gentle traction for intraoperative reduction became apparent. i Uneventful healing after 10 days.
8
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Michael Blauth
• Trophic changes: Arterial disease may result in ischemic changes and poor healing while venous hypertension produces edema, ulcers, and chronic skin changes. Using minimally invasive surgical (MIS) techniques may help to reduce problems. • Hematoma: Surgeons must take great care to lose as little blood as possible. Meticulous hemostasis helps avoid tipping the patient out of equilibrium. Subcutaneous hematoma should be evacuated even with active anticoagulation to avoid rapid skin breakdown. • Muscles are frequently atrophied and weaker than in younger patients (sarcopenia). Any manipulations during surgery should be carried out gently. Minimally invasive procedures are generally preferred.
5
Bone quality
Bony quality varies substantially from the typical wide osteoporotic tube with thin cortices to a thickened but brittle cortex in atypical fractures. Thus, cortex perforation or other iatrogenic damage generated by clamps or lag screws is more likely to occur than in normal bone (Fig 1.2-2). Forceful reduction maneuvers and aggressive handling of bone may result in extension of the injury beyond the original pattern. The use of clamps must be performed cautiously to avoid additional damage (Fig 1.2-3). Avoid the use of crushing reduction forceps helps avert worsening the comminution. Fracture patterns are often complex, with impaction occurring in the setting of a low-energy trauma. Interestingly, the impact of osteoporosis as a standalone factor on “mechanical failures” of implants could not be shown in several clinical studies. Quality of reduction and implant placement are obviously even more important [1, 2]. In a retrospective study of proximal humeral fractures, it was shown that the risk for mechanical failure increases significantly with the combination of several negative factors [3].
a
b
a
c
d
b
e
Fig 1.2-2a–e a A 76-year-old woman with a simple 2-part fracture of the left humerus. b After anatomical reduction, a 3.5 mm titanium lag screw was used to provide absolute stability (not displayed). After tightening the screw just a little bit too much, a multifragmentary situation emerged. The reduction was challenging and a bridging type of construct was chosen. c–e Uneventful healing after 2 months (c, d) and 5 months (e). The patient did not even have osteopenia.
d
e
c
Fig 1.2-3a–e a A 70-year-old woman with a humeral shaft bending wedge fracture (12B2 [14]). b Open reduction and retention with multiple clamps. c More manipulation led to a multifragmentary situation that was difficult to align and fix with a locking plate. d–e Result with excellent clinical function after 3 months.
9
Section 1 Principles 1.2 Principles of orthogeriatric surgical care
6
Bone deformation
Anterior and lateral bowing of the femur have a clinical impact in geriatric fractures and may make it very challenging to use standard intra and extramedullary implants [4]. A recent report also found that a significant increase in the lateral and anterior bow of the femur was associated with low-energy femoral shaft fractures. Therefore, the increased bowing of femoral shaft should be recognized as an important risk factor of this injury [5]. Specifically, lateral bowing of the femoral shaft may be increased in older adults as well as in younger patients with decreased bone mineralization. Osteoporosis or osteomalacia induce a varus or bowing of the femur. The lateral femoral shaft is subjected to tensile strains during a variety of physical activities; walking has the strongest impact. This effect will be pronounced with bowing in osteoporotic patients [6]. Preexisting advanced varus knee osteoarthritis, with shifting the mechanical axis medially, has been considered as a minor reason for bowing of the femoral shaft. Although atypical femoral fractures have been associated with long-term use of bisphosphonates (BPs), it was also noted that these fractures may develop without BPs use, especially in patients of Asian descent. In 2013, the Task Force of the American Society for Bone and Mineral Research revised the definition of atypical femoral fracture, removing specific diseases and drug exposures as one of the association from the minor features [7]. According to this definition, stress fractures caused by femoral bowing deformity may also be classified as atypical femoral fractures. Despite being the most commonly recommended implant choice, intramedullary (IM) nails can be difficult to insert, as the curvature of IM nail is different from that of the radius of bowed femur. In cephalomedullary nailing, the distal end of nail may break or penetrate the anterior cortex of femur in the distal segment. Reaming is often difficult as well and must be performed gently due to the narrow medullary canal and the brittle nature of the bone. Also, the nailing may cause an inadvertent fracture or malreduction with a bony gap on the medial aspect of the bone, especially in the atypical femoral shaft fractures with bowing [8]. This effect may result in impaired fracture healing or even nonunion.
10
Plate fixation can be a solution in bowed femoral fractures. In such cases, the plate may need to be contoured before fixation, considering the contralateral, noninjured leg. Otherwise, the proximal or distal end of plate will step off the bone, and it may be a source of malreduction when screws are tightened [4].
7
Classification
Classification of fragility fractures is often challenging because of different fracture patterns. Osteoporotic fractures often occur in patterns not described in the currently used classification schemes. This frustrates attempts to classify the fractures and may result in incorrect procedure or implant selection. The AO/OTA Fracture and Dislocation Classification is useful for many, but not all, fragility fractures.
8
Indications for fixation
Most fractures of the lower extremity should be surgically managed. In a small group of bedridden, terminal patients, nonoperative palliative management of hip and other lower leg fractures may be adequate. Those decisions should be team decisions made with the geriatrician, patient, family, and medical team. For the upper extremity, the need to preserve function should be considered to allow the patient to accomplish activities of daily living like eating, self-care, grooming, and ambulation. Attaining these goals may involve taking more surgical and overall risk. Therefore, surgical treatment may only be indicated if it will result in a significant improvement in function. In the proximal humerus, olecranon, and distal radius, nonsurgical management often leads to an acceptable functional result [9–11]. Some nonsurgical approaches are not tolerated as well as in younger individuals. Casts interfere with functionality and increase the risk of falls. Immobilization may render old patients immediately dependent for basic activities like eating and grooming, and promote accelerated functional decline. In a sense casts are also tethers that patients have difficulties to deal with. The cast will prevent a patient from accomplishing daily activities like walking, and the patient may therefore require placement in a nursing home. Casts and braces tend to exacerbate delirium in older adults (Fig 1.2-4).
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Michael Blauth
Complete recovery after trauma is typically the goal of treatment below the age of 60 years. This does not apply to FFPs. In this age group, we focus on the restoration of individual functional needs. Decision making can be difficult due to the variable physiological and functional nature of older patients. It is often necessary to individualize treatment approaches with the consensus of the orthogeriatric team and patients’ family.
9
10
Single shot surgery
It is obvious that any kind of revision surgery must be avoided because of the limited patient reserves necessary to tolerate and recover from surgery and functional decline. The choice of treatment should be influenced by this principle. Hemiarthroplasty instead of fracture fixation for femoral neck fractures and other primary joint replacement surgeries are good examples.
Positioning 11
Correct intraoperative positioning avoids pressure sores and skin damage: It is essential to carefully position the patient on the surgical table. Avoidance of pressure sores is of particular importance as sores significantly interfere with recovery and take an extended time to heal. An infected pressure sore may actually result in sepsis and death in the older fracture patient. In most cases, the supine position is preferred to allow for overall care by the anesthetist. When under regional anesthesia, the patient can breathe easier when supine and this position is usually more comfortable.
eight bearing as tolerated and functional W aftertreatment
Usually, the surgeon’s attention is focused on the intraoperative and immediate perioperative treatment period. Postoperatively, if the wound healing is progressing normally and x-rays are satisfactory, limited attention is paid to rehabilitation options and progress. The communication among surgeons, staff nurses, and physiotherapists regarding mobilization issues is often poor. Early postoperative mobilization and unrestricted weight bearing as tolerated are important principles for a multitude of reasons. Prolonged bed rest or “sitting mobilization” are not adequate options because of the following consequences: • Loss of muscle mass represents an independent risk factor for new falls and fractures in older adults.
a
b
c
d
e
Fig 1.2-4a–e a A 92-year-old woman with a humeral fracture (12B3). Bracing was not tolerated well. b–c After 10 days close reduction and fixation with a long multilock nail. d–e Uneventful healing after 3 months. The function reached the preinjury level.
11
Section 1 Principles 1.2 Principles of orthogeriatric surgical care
• Restriction of weight bearing inflicts a significant physiological burden on the geriatric patient. The energy expenditure for ambulation without full weight bearing increases fourfold, leading to rapid exhaustion [12]. • Fragility fracture patients are often physically unable to perform partial weight bearing due to sarcopenia, lack of proprioception and weakness in the arms; or they are admitted with an already impaired functional deficit in upper and lower extremities, preventing them from using crutches or walkers in a way that the affected lower extremity is effectively spared. • Patients develop unnecessary fear and get anxious about their inability to return to their preinjury functional status. Consequently, motivation may drop. The altered gait mechanism needs cognitive input and may lead to complaints of overload or low back pain. • Many FFPs have some degree of cognitive impairment. They may not understand (or rapidly forget) instructions and instead follow their own impulses. • Partial weight-bearing protocols are not evidence-based but often the result of the surgeon’s own uncertainty. • Even for patients on adequate pain medication, pain will typically guide the patient to use the appropriate weight bearing and safely progress with ambulation. Patients with severely impaired cognitive function are more prone to fall, but they have the same self-protective mechanisms as cognitively normal patients.
a
b
c
Early weight bearing can promote fracture healing and union of the fracture without increasing loss of fixation [13, 14]. Immobilization of joints is poorly tolerated in many older patients; early functional range of motion prevents joints from stiffening. The daily loss of muscle mass during periods of bed rest is dramatic. Modern surgical procedures and implants permit immediate unrestricted weight bearing for most fractures. Temporary external transarticular fixation can be a unique solution in fractures around the knee if internal fixation does not seem to be stable enough for immediate mobilization, if soft tissues have to settle down or if there is no chance to apply implants directly to the bone (Fig 1.2-5) [15].
12
Fixation techniques
The major technical problem the surgeon faces is the difficulty producing secure fixation of the implant to the bone. There is less cortical and cancellous bone for the screw threads to engage and the pullout strength of implants is significantly lower in osteoporotic bone. Bone mineral density correlates linearly with the holding power of screws. If the load transmitted at the bone-implant interface exceeds the strain tolerance of osteoporotic bone, microfracture and resorption of bone with loosening of the
d
e
Fig 1.2-5a–e a–b A 75-year-old woman with low periprosthetic fracture after total knee arthroplasty (TKA) and severe comorbidities. c Temporary transarticular fixation for 8 weeks. d–e Bony healing after 3 months. Final range of motion 0–10–100°.
12
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Michael Blauth
implant and secondary failure of fixation will occur. The common mode of failure of internal fixation in osteoporotic bone is bone failure rather than implant failure.
The general principles of fracture management are applicable to most fragility fractures, but the decrease in bone strength requires some adaption to decrease the risk of failure.
gap must be closed as much as possible, ie, bone contact must be achieved. Three to four holes should be left free and three to four bicortical locking head screws (LHSs) in each main fragment are needed. • Spiral-type 2-part fractures should be reduced and “adapted” as much as possible and preliminarily fixed with suture or hardware cerclages or cables. If screws are used, they should be tightened with caution as “reduction screws”. The first plate screw should be inserted at the end of the fracture line. Three to four bicortical LHSs in each main fragment are necessary depending on the type of bone (Fig 1.2-6). • In comminuted fractures, the first screws should be placed adjacent to the fracture zone. Four to five bicortical screws in each main fragment are sufficient.
12.1 Minimally invasive surgery
12.3 Splinting the whole bone
Minimally invasive surgery (MIS) techniques feature multiple “traditional” advantages that are even more helpful in FFPs than in younger patients. Many older adults are anticoagulated and suffer already from muscle weakness. Technically, MIS is easy to perform as soft-tissue layers can be separated easily. For more details, see Blauth et al [16].
Subsequent fractures adjacent to the end of plates, nails or prosthesis occur due to the stress riser between the stiff implant and the soft bone. The frequency is not clear. If possible, the whole bone should be protected at the first fixation including the femoral neck in case of the femur (Fig 1.2-7, Fig 1.2-8). To achieve this goal, sometimes a combination of intramedullary with extramedullary implants becomes necessary.
Internal fixation must take the local bone mineral distribution into account. This varies with fracture location, age, and gender. Proper preoperative planning, implant choice, fixation technique, and understanding of the biomechanical principles are essential.
Specifically designed instruments for MIS are available. It is important to develop a familiarity with their use.
12.4 Angular stable implants and blades 12.2 Relative stability
Thin cortices cannot withstand the compressive forces that are needed to create absolute stability. Tightening lag screws a little too much may create iatrogenic fractures that worsen the situation significantly (Fig 1.2-2, Fig 1.2-3). In osteoporotic bone it may not always be possible to obtain and maintain anatomical reduction and compression with absolute stability because the weakened cortical and cancellous bone may fail under compression. It is essential not to mix the principles of relative and absolute stability in one fracture fixation. As a simple rule, intramedullary devices are preferred over extramedullary devices if fracture patterns and soft tissues allow for it. Unfortunately, for metaphyseal fractures around the knee, locking options are not yet optimized for osteoporotic bone and thus nails are often not applicable. Short plates with every screw hole filled will cause concentration of forces, which may exceed the strain tolerance of osteoporotic bone. Basic rules have been previously established in the literature [17, 18]: • Simple transverse fractures are best addressed by intramedullary implants. If this is not possible, the fracture
Implants with locking head mechanism and fixed or variable angle between screw and plate as well as angular stable locking options for intramedullary nails all have biomechanically shown to provide superior stability in bone with reduced cortical thickness. Locking head screws cannot be overtightened or overinserted rendering them unstable because the thread gets destroyed. They should always be used in a bicortical mode to improve their working length with thin cortices. In addition, locking screws have a larger core diameter than conventional screws, which results in a higher pullout strength and overall strength. This is especially helpful in metaphyseal bone where intramedullary nails may fail. The holding power of the LHS can further be increased by orienting them in different directions: This method is used with the proximal humeral plate and the distal femoral and proximal tibial plates. A blade for fixation of pertrochanteric fractures offers biomechanical advantages over a lag screw. The blade condenses the bone around the implant, while screw insertion always results in some bone loss.
13
Section 1 Principles 1.2 Principles of orthogeriatric surgical care
a
b
c
d
e
f
Fig 1.2-6a–f a–b a A 77-year-old woman with a pertrochanteric fracture (31A2). b Fixation with proximal femoral nail antirotation. c–d The nail was removed 1.5 years later because of lateral thigh pain. Three years later, she sustained a spiral diaphyseal fracture (32A1). e–f Minimally invasive reduction in lateral position and preliminary fixation with suture wire. Definitive fixation in relative stability with distal femoral plate, the first proximal screw starting at the end of the fracture and 10 cortices. Uneventful healing with small callus formation. Ideally, a longer plate to protect the whole femur would have been indicated.
a
b
c
Fig 1.2-7a–c a A 92-year-old woman with periprosthetic fracture type B2. b–c Open reduction, fixation with cerclage wires and revision arthroplasty with a long-stemmed implant with locking options. Distal femoral plate to protect the bone between the two prostheses.
14
a
b
c
e
f
g
d
h
Fig 1.2-8a–h a–b An 80-year-old woman with a periprosthetic knee fracture. c Two and a half months after fixation with a distal femoral plate fracture adjacent to the proximal end of the plate. d Application of a longer plate. Fixation in varus malalingment and with the fracture gap still open. e The construct is too stiff and fails after another 2.5 months. f–h Final solution with antegrade femoral nail. Distal locking with axial loading screws.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Michael Blauth
12.5 Anatomical alignment
12.7 Augmentation with polymethylmethacrylate
Correct anatomical alignment represents an important prerequisite for uneventful bone healing. Fixation of osteoporotic bones is less tolerant for any deviation than in younger bone. Specifically varus malalignment should be avoided in femoral fractures.
Fixation in osteoporotic bone can be improved by augmenting the bone with cement. Augmented purchase of the implant, in particular of screws, reduces the risk of hardware migration, cut out, cut through and pull out. It can also be used as a void filler to support the bone structure, for example, of a vertebral body or the tibial plateau, and prevent it from collapsing.
Severe rotational malalignment is an underrecognized problem and occurs typically with very unstable proximal femoral fractures. Rotational malalignment should be avoided.
Polymethylmethacrylate (PMMA) remains the material of choice and may be used in different ways:
12.6 Bone impaction
• For filling voids that mainly result after reduction of cancellous bone. A typical example is vertebral body compression fracture treated with closed reduction with vertebroplasty or kyphoplasty. The same principle can be applied to proximal tibial fractures; cement used as a void filler prevents the articular surface from collapsing after elevation. • In standardized implant augmentation, the cement is typically injected with a specific cannula through perforated implants to improve the bone-implant interface by preventing high bone strain and distributing the force to the bone in a load-sharing rather than load-bearing configuration (Fig 1.2-9). • In nonstandardized implant augmentation, the cement is applied via the screw hole or cortical window before or after the implant is inserted.
Bone impaction at the fracture site is a key element in the surgical management of osteoporotic fractures as it reduces the risk of implant failure. In many cases, like for example in the valgus-impacted fracture of the femoral neck, impaction is created by the trauma itself. Controlled impaction can be attained by tensioning internal fixation devices. Implants, such as the dynamic hip screw, which allow for controlled impaction of the fracture while preventing penetration of the joint by the hip screw.
a
b
c
d
Fig 1.2-9a–d a An 82-year-old man with a proximal femoral fracture (31A2). b Close reduction with traction table. After insertion of nail and blade, the decision was taken to augment the blade because of severe osteoporosis and a very low resistance while inserting the blade. Intraoperative contrast dye test demonstrated no arthrogram, ie, no perforation into the hip joint. c–d Injection of 4 mL of polymethylmethacrylate through a special cannula. Result after mobilization with center-center position of the headneck-element and equally distributed cement.
15
Section 1 Principles 1.2 Principles of orthogeriatric surgical care
Standardized implant augmentation has been thoroughly studied in recent years: • Many sites have been tested biomechanically. In the proximal femur, proximal humerus, proximal tibia and sacrum, augmentation with PMMA cement improved cycles required to cause mechanical failure by ~ 100%; this applies only in osteoporotic bone. • Small volumes of cement are sufficient. Larger quantities do not improve implant purchase significantly. • Heat generation outside the cement does not exceed 42° C, because the metallic implant serves as a heat sink for the exothermic chemical reaction. • No signs of cartilage damage next to the cement mass were noted in sheep experiments. • Interference with bone healing has not been demonstrated so far.
12.8 Autografts
Corticocancellous bone autografts to assist fracture healing and to fill gaps can also be harvested in older patients. Unless used as avoid filler, grafts should be fixed to the bone by cortical screws (Fig 1.2-10). 12.9 Allografts
Allograft bone has good mechanical properties but less osteogenic potential compared to autografts. In osteoporotic bone, allografts are used to fill metaphyseal voids and to prevent articular and other fragments from subsiding. This can be helpful in fractures of the proximal and distal humerus, distal radius and proximal tibia. Allograft struts are also used in periprosthetic femoral fractures with poor bone quality to enhance the mechanical strength of the construct (Fig 1.2-11).
Standardized implant augmentation with PMMA limits the negative effect of osteoporosis on implant fixation, “converting” osteoporotic bone into normal bone.
a
d
16
b
e
c
f
g
Fig 1.2-10a–g a–c A 70-year-old woman with an unstable 3-part fracture. d–e Fracture fixation was indicated despite the obvious risk for avascular necrosis because a stable reconstruction seemed to be possible. Anatomical reduction and fixation with PHILOS. f S tandardized implant augmentation via cannulated locking head screws with 0.5 mL of polymethylmethacrylate each to minimize the risk of mechanical failure. g Injection of cement is only indicated and possible in osteoporotic bone. Follow-up after 3 months.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Michael Blauth
12.10 Joint replacement
Joint replacement plays an important role in older patients. It is commonly used in the proximal femur, mainly with femoral neck fractures. The indication for fracture arthroplasty is not as clear in proximal humeral fractures. A reverse shoulder arthroplasty is useful in cases where stable fixation is not possible. The use of an endoprosthesis in fractures of the distal humerus, distal radius and proximal tibia remains controversial.
a
b
d
e
More rapid restoration of adequate function along with a reduced life expectancy and fewer revision surgeries are appealing arguments in favor of immediate joint replacement. There is a paucity of published evidence to inform clinical care in this area. If the general goals of fracture treatment can be achieved without violation of the above-mentioned principles, fracture fixation is usually preferred.
c
f
g
h
Fig 1.2-11a–h a–c A 76-year-old woman with a displaced 2-part fracture of the proximal humerus. Severe osteoporosis with T-score lumbar spine -3.8, femoral neck -3.6 and a slender head fragment. d–f Central void after open reduction (d) that is filled with a structural allograft from the bone bank (e–f). g–h Follow-up after 3 months.
17
Franz Kralinger, Michael Blauth
3.1 Proximal humerus Franz Kralinger, Michael Blauth
1
Introduction
The treatment of proximal humeral fractures (PHFs) is controversial for a number of reasons: • There is an ongoing debate about the benefits of operative versus nonoperative treatment even of displaced, unstable fractures. • The rate of “mechanical” complications after surgical treatment is 30–35% in prospective studies with surgical revision rates from 20–30% [1–3]. Nevertheless, up to 74% of PHFs actually received surgical treatment [4]. • Multiple available operative options from pinning to arthroplasty with varying selection criteria are mainly based on bone and fracture characteristics without considering the patient’s functional status. • There is a lack of randomized studies investigating distinct fracture entities and treatment modalities. • Unspecific outcome measures. While using the Constant and Disabilities of the Arm, Shoulder, and Hand (DASH) scores, ceiling effects may make it difficult to detect substantial advantages of surgical fixation over nonoperative management [5]. If the scores or patient-reported outcome measures (PROMs) used were more precise, differences between different procedures might be detected more easily. A ceiling effect of a score describes the fact that a score at its upper end of values loses the ability to detect changes in a patient’s health status in a sufficient manner. Therefore ceiling effects can lead to artefactual data, eg, showing no effect of an intervention when in reality there is one. • More traditional ways to evaluate treatment results like range of motion (ROM), muscle strength, fracture reduction, and bone healing may not apply to the proximal humerus where objective parameters often do not match subjective appraisals [6].
Since there is no single clear-cut approach to a PHF, treatment recommendations depend largely on surgeon experience, skills, and preference. An improvement of this situation can only be achieved with larger and higher level clinical studies and specifically designed PROMs to address the geriatric population. This chapter summarizes the current situation from a practical approach to guide proper patientand treatment-specific decision making. This seems to be the most important factor in achieving a good outcome as well as for the most appropriate treatment selection and avoidance of complications. 1.1
Epidemiology
Proximal humeral fractures: • Are the third most common fractures in adults older than 60 years. • Affect 70–80% of women with a history of osteoporosis who have fallen from a standing height [1]. • Have been increasing by up to 15% in the past decades [7, 8]. There are major differences between ethnicities and the rate of fractures are significantly lower in countries like Japan [9] compared to Europe or America. Interestingly, the clear rise in the rate of low-trauma PHFs in older Finnish women from the early 1970s until the mid1990s has stabilized at a high level. The reasons for this are largely unknown, but a cohort effect toward a healthier aging population with improved functional ability and reduced risk of injurious falls cannot be ruled out [10]. In Austria, on the other hand, this levelling off effect could not be confirmed and absolute numbers of PHFs are still rising due to increased life expectancy [11].
191
Section 3 Fracture management 3.1 Proximal humerus
1.2
Etiology
CASE 1
• Proximal humeral fractures occur mostly after low-energy falls [12]. • Comorbidities increase the risk for PHFs. Factors like decreased neuromuscular response, delayed reaction time, cognitive impairment, impaired balance, intoxication as well as early menopause are all associated with PHFs [13]. • Middle-aged patients who sustain PHFs are physiologically older than their numerical age indicates and have a higher incidence of medical comorbidities often related to alcohol, tobacco, and drug usage (Case 1: Fig 3.1-1) [14].
Young woman suffering from alcoholism and severe osteoporosis, multiple fractures including bilateral proximal humeral fractures with special solution on one side Patient A 48-year-old woman with no obvious signs of dementia or confusion; she was cooperative. She was living with her husband and wanted to remain independent. Over the years, she sustained continuously major fractures of the distal radius, lumbar spine, and proximal tibia. Comorbidities • Alcohol addiction • Nicotine abuse • Chronic obstructive pulmonary disease • Osteoporosis • Grand mal seizures • Multiple other comorbidities Treatment and outcome In 2005 the female patient sustained a subcapital fracture of the left humerus (Fig 3.1-1a–c). The fracture was treated nonoperatively leading to healing in malalignment. The bone was clearly osteoporotic, but no action was taken with regard to this. In 2008 she presented with a fracture of the right proximal humerus (Fig 3.1-1d–f). The fracture was first treated nonoperatively, which resulted in a painful and debilitating condition. Since fracture healing was not to be expected, the decision to perform surgery was made.
The intraoperative (Fig 3.1-1j–k), 3-month (Fig 3.1-1l), and 1-year (Fig 3.1-1m) follow-up x-rays showed an uneventful clinical course. Range of motion was 120° of abduction/flexion and 60° of external rotation. Treatment options • Nonanatomical fixation in valgus of the humeral head and massive shortening may result in an impaired functional outcome due to a shorter lever arm of the rotator cuff muscles. • Nailing: Head fragment is too short for stable anchorage of the fifth anchor point. • Hemiarthroplasty: Overtreatment when stable reconstruction is possible. Midterm function of the shoulder is questionable. • Reverse total shoulder arthroplasty is not indicated with intact rotator cuff and well-centered shoulder joint. Key points • Central voids can be successfully filled with massive allografts to prevent early varus failure and subsidence of the head fragment with cut-through of the screws, even in patients with severe osteoporosis [15]. • In a retrospective case series, this procedure leads to bony union in a noncompliant or high-risk patient population [16]. • Treatment of the underlying osteoporosis may be challenging in noncompliant patients.
After the fracture was aligned (Fig 3.1-1g), an almost normal anatomical condition could be reestablished with a massive central allograft (Fig 3.1-1h). Then the plate was preliminarily fixed (Fig 3.1-1i).
192
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
a
g
b
h
c
i
d
j
e
k
f
l
m
Fig 3.1-1a–m A 48-year-old woman with multiple fractures. a–c X-rays showing a subcapital fracture of the left humerus. d–f X-rays showing a fracture of the right proximal humerus. g–i X-rays showing aligned fracture ( g), an almost normal anatomical condition achieved with a massive central allograft (h) and preliminary plate fixation (i). j–m Intraoperative result ( j–k), follow-up after 3 months (l), and 1 year (m). Note the different projection in the last picture which can be read from the PHILOS plate.
2
Diagnostics and classification
2.1
Clinical evaluation
History and physical examination include: In order to give a viable therapeutic recommendation, the preoperative workup must go beyond fracture analysis, although there seems to be a high degree of uncertainty as how to measure and implement clinical information into the decision-making process. In a recent study 238 surgeons rated 40 x-rays of patients with PHFs. Participants were randomly selected to receive information about the patient and mechanism of the injury. Patient information, particularly older age, was associated with a higher likelihood of nonoperative treatment recommendation rather than x-rays alone. Clinical information did not improve agreement with the actual treatment or the generally poor interobserver agreement on treatment recommendations [17].
• Mechanism of injury • Vascular and neurological status, especially distal circulation and the axillary nerve function • Soft-tissue injuries, including the skin • Muscle status, specifically the muscles of the rotator cuff (Diagnostics 1: Fig 3.1-2, Fig 3.1-3, Fig 3.1-4) • Preinjury level of function • Occupation • Hand dominance • History of malignancy • History of previous fragility fractures • Rehabilitation potential • Presence of concomitant injuries • Geriatric workup including comorbidities, functional and mental status
193
Section 3 Fracture management 3.1 Proximal humerus
DIAGNOSTICS 1
Evidence of how the mental status may influence the outcome is poor. In most studies, patients with significant mental impairment are excluded or this factor is not considered at all. Advanced age and higher degrees of dementia with increased risk of postoperative delirium usually lead to nonoperative treatment (Case 2: Fig 3.1-5).
Other patient factors like level of independence, housing situation, or the need to use walking aids also potentially affect outcomes after both operative and nonoperative management and should therefore be evaluated very carefully. Complications such as infection, nonunion, osteonecrosis, fixation failure, and compliance with rehabilitation can all be related to medical comorbidities [13]. Alcohol abuse particularly increases a patient’s risk of noncompliance and nonunion, and tobacco use increases the risk of nonunion [18].
Evaluation of rotator cuff muscles status with standardized 2-D computed tomographic reconstructions Patients An 85-year-old woman with chronic rotator cuff deficiency and complete atrophy of the supra- and infraspinatus muscles (Fig 3.1-2). A 78-year-old woman with posterior and superior cuff deficiency (Fig 3.1-3).
Discussion A fracture reconstruction in a patient like the one in Fig 3.1-3 does not seem to be indicated. Even if the pretrauma status of the computed tomographic angiography was compensated, the risk of decompensation after the reconstruction with the need of revision surgery is high; the authors recommend reverse shoulder arthroplasty in these cases.
An 87-year-old man with no muscle atrophy or fatty degeneration of the rotator cuff muscles (Fig 3.1-4).
a Fig 3.1-2 Clinical picture of an 85-year-old woman with chronic rotator cuff deficiency and complete atrophy of the supraspinatus muscle and infraspinatus muscles. The diagnosis can be easily made by visual inspection.
b
c
Fig 3.1-3a–c Parasagittal 2-D reconstruction of a 78-year-old woman with posterior and superior cuff deficiency. Inhomogeneous presentation of the supraspinatus muscle and infraspinatus muscle because of atrophy and fatty degeneration (a, c). The subscapularis (SSC) is still in good shape in the caudal aspects (b).
a
b
c
Fig 3.1-4a–c In contrast, parasagittal 2-D reconstruction of an 87-year-old man shows no muscle atrophy or fatty degeneration of the rotator cuff muscles. Note the muscle belly of the supras pinatus muscle (a, c). The head fragment is internally rotated due to the pull of the subscapularis tendon, and the corresponding muscle is without atrophy (b). The full cuff without significant atrophy or fatty
infiltration is visible in the parasagittal plane at the coracoid and base of the spinal junction [19].
194
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
Patient A 90-year-old slow-go, ie, unfit, female patient, was living in a nursing home and required a walker to assist with ambulation. Comorbidities • Dementia • Coronary heart disease • Hypertension • Multiple falls • A pertrochanteric fracture 2 years ago, treated with a cephalo medullary nail with an augemented head neck element or so-called proximal femoral nail antirotation plus augmentation • Osteoporosis Treatment and outcome The female patient had a displaced 2-part proximal humeral fracture (Fig 3.1-5a–b) and an undisplaced superior and inferior anterior pelvic ring fracture (Fig 3.1-5c). Her therapy comprised shoulder sling, pain medication, and pain-adapted mobilization. She was hospitalized for 16 days, mobilized to sit in a wheelchair, and transferred to a nursing home (Fig 3.1-5d). The follow-up x-rays at 6 weeks showed a rapid ongoing healing process in varus malalignment (Fig 3.1-5e–g). She had only little
a
e
b
f
c
g
CASE 2
Evaluation of mental status and comorbidities
pain and could reach her head with her hand. She did not walk any more but was satisfied with her situation and refused further followups. Her situation was still the same 2 years later. Discussion • From a geriatric standpoint, everything should be done to get the patient out of bed: adapt pain medication, keep motivating her and help her to walk again. Bed rest with loss of muscle mass, staring at the ceiling all the time and eating in bed has to be avoided. Nutritional aspects should also be considered. • To answer the question if surgical treatment under a nerve block would improve the patient’s prognosis, her prefracture status needs to be carefully evaluated (ie, “What was she really able to do?“) as well as her mental status, ability to cooperate, and motivation. This may take a few days. Finally her risk for surgery must be estimated. • In this case, despite a low risk for surgery, nonoperative treatment was recommended, because the patient had poor cognitive function and did not require high functional demands. In her case bone healing took place quickly despite her severe osteoporosis. Surgical stabilization would most probably not have caused any change in the rehabilitation process. • From a surgical standpoint, if a nailing procedure would have been chosen, a low risk for failure would have been expected.
d
Fig 3.1-5a–g A 90-year-old woman with a 2-part fracture of the proximal humerus. a–b X-ray of a displaced 2-part proximal humeral fracture. c X-ray showing the undisplaced superior and inferior anterior pelvic ring fracture. d T herapy comprised shoulder sling, pain medication, painadapted mobilization. e–g Postoperative follow-up x-rays after 6 weeks showing a rapid ongoing healing process in varus malalignment.
195
Section 3 Fracture management 3.1 Proximal humerus
To describe the functional status of the patient, a simple distinction between go-go, slow-go and no-go patients (Table 3.1-1, see topic 3.1 in this chapter) is useful. The P arker Mobility Score and the WHO performance status (Table 3.1-2, see topic 3.3 in this chapter) may also be helpful. 2.2
Imaging
2.2.1 Plain x-rays
Surgeons can only make clear and unambiguous statements if the fracture is clearly visualized by x-ray(s). If criteria are defined to classify and treat PHFs, the proximal humerus must be displayed in a manner that those criteria can be reliably evaluated. The trauma series consists of a true AP view, an axial view, and an outlet view. The first two x-rays are most important to check the displacement of fragments and the instability of the fracture. Acute pain should be treated before images are taken. Analyzing the projection of the proximal humerus in published serial x-ray studies suggests that the position of the patient’s arm often varies within one case. Recommendations about the “standard position” also vary widely. According to geometrical studies by Hengg et al [20], especially different degrees of internal rotation (IR) distort the measurement of the head-shaft angle on the AP view substantially: 30°, 45°, and 60° of IR result in a projection of the head-shaft angle of 144°, 150°, and 159°. Standardized and above all comparable visualizations of the proximal humerus are therefore crucial to make decisions and to collate results (Diagnostics 2: Fig 3.1-6).
The axial view (Diagnostics 4: Fig 3.1-9, Fig 3.1-10): • Is paramount to assess anterior or posterior displacement of the humeral head in relation to the glenoid • Determines anteversion and retroversion • Displays displacement and fragmentation of the GT (Diagnostics 5: Fig 3.1-11) and the overlap of the head by the minor tuberosity • Shows posterior dislocation of the humeral head associated with PHF, which is often missed without an appropriate axillary lateral view. Alternatively, a dynamic investigation under image intensifier may be performed The Velpeau view is an alternative to the axial view and can be obtained with the arm in a sling. The lateral view (= lateral scapula, = Y view, = outlet view) (Diagnostics 6: Fig 3.1-12, Fig 3.1-13) is easy to shoot in the trauma situation but often very difficult to interpret because of poor quality and superimposed structures. It is definitely the third most important view of the trauma series. If only two views are done, they should be the true AP and axial views. The lateral view shows: • Greater tuberosity posterior displacement due to the pull of the infraspinatus (ISP) and supraspinatus (SSP) muscles. • The relation of the head fragment to the glenoid.
The true AP view (Diagnostics 3: Fig 3.1-7, Fig 3.1-8) shows: • Varus and valgus deformity and amount of displacement • Medial displacement of the shaft consistently produced by the pectoralis major muscle • Posterosuperior displacement of the greater tuberosity (GT) Rotational displacement of the head fragment is due to the pull of the subscapularis (SSC) in 3-part GT fractures. This pathology needs to be derotated in percutaneous procedures. When placing the arm in a sling or holding it in a relieving posture, this type of view cannot be achieved.
196
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Comparable projections are of utmost importance
a
b
c
d
Fig 3.1-6a–d X-ray series after a displaced 3-part fracture in a 58-year-old woman (a). In a true AP view taken postoperatively the head-shaft angle (HSA) amounts to 135°, which is equivalent to an anatomical reduction (b). The 8-week follow-up displays varus malalignment of 120.9° which is due to only another arm rotation and also an x-ray beam projection. This can be easily detected by comparing the projection of the standard locking plate in both views (c). The 1-year follow-up shows again the initial situation with a HSA of 133° and no relevant loss of reduction (d). This example clearly demonstrates the great importance of comparable standardized projections.
DIAGNOSTICS 3
True AP view
a
a
DIAGNOSTICS 2
Franz Kralinger, Michael Blauth
b
b
Fig 3.1-7a–b The patient’s affected shoulder should be placed against the x-ray plate with his trunk tilted approximately 40° toward the beam. The scapula of the affected shoulder should be parallel to the cassette (a). The patient’s arm is in neutral rotation, ie, with the thumb bent forward; this position is reproducible and complies with the geometrical reflections of a true AP view (b). The central beam is orientated 20–25° caudally.
Fig 3.1-8a–b Orthograde and tangential projection of the glenoid, and free projection of the humeral head with the greater tuberosity (GT) marginalized and the subacromial space visible. Examples of an uninjured shoulder of a 32-year-old man without pathology (a) and a 3-part GT valgus-impacted proximal humeral fracture in a 42-year-old woman (b), both in correct AP view.
197
Section 3 Fracture management
DIAGNOSTICS 4
3.1 Proximal humerus
Axial view
a
b
c
d
e
f
Fig 3.1-9 Axial view in 30–40° of abduction and the forearm parallel to the table. This can be achieved in most acute cases after administering some pain medication.
h
k
198
i
g
j
Fig 3.1-10a–k A 72-year-old woman with a 2-part fracture. a–c True AP, outlet, and axial views of a displaced 2-part fracture in a go-go, ie, fit, 72-yearold female patient. Displacement and instability is best demonstrated in the axial view. d–e Fracture fixation with an intramedullary nail. f–g Result after 6 months. Note that the projection is different from the postoperative views. h–k Functional rehabilitation 8 days after surgery.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
DIAGNOSTICS 5
Franz Kralinger, Michael Blauth
a
c
b
Fig 3.1-11a–c The AP and outlet views (a–b) provide sufficient information about the greater tuberosity (GT), shaft and head fragment. The axial view (c) displays all missing information to classify the HL-G-S fracture. Note the comminution and dorsal displacement of the GT.
DIAGNOSTICS 6
Lateral view
a Fig 3.1-12 For the lateral view, the anterior shoulder is placed on the x-ray plate with the unaffected shoulder tilted forward 40°. The beam is placed posteriorly and directed along the scapular spine.
b
Fig 3.1-13a–b The AP view (a) of an 80-year-old patient shows a fracture involving the greater tuberosity, the shaft, and the head fragment. The lateral view (b) does not add much information. Especially the lesser tuberosity (LT) cannot be visualized well, be it in the AP view or in the lateral view. The axial view generates this critical information, and if unavailable, a computed tomographic scan is necessary to show the involvement of the LT.
199
Section 3 Fracture management 3.1 Proximal humerus
2.2.2 Computed tomographic scan
CASE 3
In most hospitals, a CT scan is part of the standard workup of PHFs. Certainly, if surgical treatment is an option, a CT scan with 3-D reconstructions adds important information in fracture dislocations, humeral head-split fractures, and comminuted fractures. Three-dimensional CT scan reconstructions have been shown to provide the highest interobserver agreement with regard to classification and treatment recommendations among upper-extremity specialists [21].
CT scans: • Help to precisely determine fracture lines and pieces, ie, fracture characteristics needed for surgical planning. • Provide enhanced understanding of fracture comminution, impaction, humeral head involvement and its size and remaining thickness, and additional glenoid articular surface injury (Case 3: Fig 3.1-14). • Allow for manual 2-D reconstructions along the axes of the humerus to show the exact angulation and displacement of fragments as well as the length of the metaphyseal fracture extension. • Facilitate soft-tissue imaging specifically rotator cuff muscles. In case of a rotator cuff arthropathy with limited preoperative function and an indication for surgery, an inverse prosthesis instead of fracture fixation is indicated.
Importance of a detailed computed tomographic analysis Patient An 80-year-old woman sustained a low-energy proximal humeral fracture and a distal radial fracture of the left upper extremity. Comorbidities • No relevant comorbidities besides osteoporosis—already treated Treatment and outcome The AP and lateral views showed a 2-part fracture of the surgical neck (HGL-S) and the medial calcar seemed comminuted (Fig 3.114a–b). Both fractures were initially treated nonoperatively. After 10 days, progressive tilting of the head fragment with pronounced displacement was visible in the lateral view (Fig 3.1-14c–d). In addition, the patient was unable to participate in rehabilitation because of pain. The surgeon and patient decided on plating. The 2-D computed tomographic scans showed the narrow head fragment (Fig 3.1-14e–f). Only very few thread pitches of the locked screws could be anchored.
Discussion In cases such as shown in Fig 3.1-14, a sustainable medial bone contact can only be achieved by reducing the head fragment into a slight valgus position. Because the length of the proximal humerus is then reduced, the plate must be placed in a nonanatomical position, ie, with a gap between lateral cortex and the plate. An impaction with valgus position and with additional cement augmentation of the screws might have helped, but if in doubt a structural allograft could definitely provide the mechanical stability for a functional rehabilitation as desired in orthogeriatrics. In a surgical neck 2-part fracture we would not consider arthroplasty.
Intraoperative x-rays demonstrated a residual varus position of the head fragment and a lack of medial support due to “wrong impaction” (Fig 3.1-14g–h). Insufficient medial support and residual varus together with the small osteoporotic head fragment led to a mechanical varus failure. After 4 weeks (Fig 3.1-14i) and 6 months (Fig 3.1-14j–k) the fracture was malunited with a severe varus deformity; since the screws did not perforate, the patient was not revised.
200
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
a
b
c
e
i
d
g
f
j
k
2.2.3 Magnetic resonance imaging
Magnetic resonance imaging (MRI) adds little to the initial evaluation of PHFs [22]. 2.2.4 Local bone quality
• Promising attempts have been made to experimentally measure the local bone quality (LBQ) with a torque measurement tool (DensiProbe) which was adapted to a stan-
h
Fig 3.1-14a–k An 80-year-old woman with a surgical neck fracture. a–b AP and lateral views showing a 2-part fracture of the surgical neck with comminuted medial calcar. c–d Lateral view showing progressive tilting of the head fragment with pronounced displacement. e–f The 2-D computed tomographic scans showing the narrow head fragment. g–h Intraoperative x-rays showing a residual varus position of the head fragment. i–k Postoperative x-rays after 4 weeks (i) and 6 months ( j–k) showing malunited fracture with severe varus deformity.
dard locking plate. The mechanical peak torque correlates with the local bone mineral density (BMD) and screw failure load in anatomical specimens [23]. • With modern picture archiving and communication systems, the local bone density (LBD) can be measured in standardized regions of interest given in Hounsfield units [24] and converted into BMD values. The aforementioned measurement gives an estimate of the LBQ.
201
Section 3 Fracture management 3.1 Proximal humerus
The significance of local osteoporosis for the outcome of the treatment is unclear. In a multicenter trial, patients with mechanical complications after plate fixation of unstable PHFs had the same low BMD as patients with uneventful healing [1]. Experiments with anatomical specimens have shown that periimplant polymethylmethacrylate (PMMA) cement augmentation cannot be injected through a cannulated screw into cancellous bone with normal density [25, 26]. If implant augmentation is intended, the LBQ needs to be determined. There is a linear biomechanical correlation between LBQ and cycles to failure [27, 28]. Common sense dictates that osteoporosis is clinically associated with increased rates of comminution and defects due to impaction, loss of fixation and reduction after surgical management. Recent clinical trial results suggest that LBQ constitutes at most one contributing factor to fixation failures after plating [1, 28].
Instability and displacement
Instability and displacement are often used as criteria for determining the treatment strategy. Imaging usually only shows a momentary situation of unstable fractures. Whether or not fracture fragments are displaced may depend on the position of the arm while x-rays or CT scans were taken. If in doubt, perform repeat x-rays examinations to help rule out misinterpretation or secondary displacement. Signs for stability are [21, 30]: • • • • • •
Minimal comminution Three or fewer fragments Absence of significant tuberosity displacement Cortical contact Relative impaction of the shaft into the head No history of dislocation
For the typical patient with PHF, LBQ must be expected [1].
If the fracture is stable, gentle and careful movements of the affected arm can be performed with no or very little pain during a physical examination. This should only be done after imaging, though.
2.3
Signs of instability are:
Soft-tissue injuries
With fractures of the GT or LT, the rotator cuff is essentially nonfunctional, as expected [13]. Conversely, we may presume that a 4-part PHF only occurs with an intact rotator cuff. Without a functioning rotator cuff, displaced avulsion fractures are rare due to the lack of pulling forces. With a preexisting cuff arthropathy, 2-part fractures are more likely. A complete rotator cuff examination cannot usually be performed in an acute setting due to pain and swelling, but the rotator cuff function should be monitored throughout the typical clinical course to ensure adequate function [22]. Due to the age of most patients who sustain PHFs previous rotator cuff injuries are likely, and a new rotator cuff tear can certainly occur in conjunction with PHFs [29]. As an indirect measurement, the status of the rotator cuff muscle can be determined with the CT scan (Diagnostics 1: Fig 3.1-2, Fig 3.1-3, Fig 3.1-4).
202
2.4
• Significant displacement with segments angulated more than 45° or displaced more than 0.5–1 cm from their normal anatomical position, best detected on the axial view [31]. • A difference in fragment angulation between plain x-rays and CT scans with 2-D reconstruction along the axes of the humerus. • Extraordinary pain which does not subside with adequate pain medication within a few days. 2.5
Classification
Codman’s 4-part model laid the foundations of modern understanding of PHFs. All of the following classifications were based on the four parts, ie, the shaft, GT, LT, and the head fragment. The most common classifications used over the last decades were the Neer and the AO/OTA classifications. Both systems are characterized by a poor interobserver reliability [32, 33] which improves with advanced imaging like 3-D CT scans [34], education and experience [35].
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
2.5.1 Neer’s classification
Neer focused on the patterns of displacement rather than the location of fracture lines. In his retrospective study he attempted to identify fractures that would benefit from open reduction. Similar to Hertel, he also wanted to predict the risk of avascular necrosis (AVN) which again would have an impact on decision making (see topic 2.5.2 in this chapter). Neer’s system remains the most commonly used today, because it is easy to apply and yet has a prognostic value. Four-part fractures generally have worse outcomes than 2- and 3-part fractures regardless of the treatment. Neer randomly defined the borderline between displaced and nondisplaced at a displacement of 1 cm and an angulation of 45°. A fracture that is below this threshold is called 1-part fracture irrespective of the number of fragments. These criteria have evolved to make a displacement of 5 mm or more an acceptable indication for fixation provided the direction of displacement creates a functional limitation. A good example is the superior displacement of the GT, which has the potential of restricting abduction.
therefore indicates the classic 4-part fracture. The letter “d”(dislocation) as a prefix to “H” and “c” followed by the length of the intact calcar fragment in millimeters as postfix in brackets can be added as well as “a” for the head-neck angulation. The simplicity and intuitive nature of this nomenclature may be the reason for a higher reliability compared to the original Hertel, AO/OTA, and Neer systems. Prediction of head necrosis does not play an important role in decision making in geriatric patients. Fracture pattern interpretation mainly serves to differentiate stable from unstable fractures and to forecast the likelihood of achieving a stable fixation. 2.6
Summary
Clinical evaluation: • In addition to fracture pattern analysis, the patient’s functional and cognitive status must be considered to determine the best approach. • Nonfracture-related geriatric parameters play an important role in choosing the adequate therapeutic approach for each individual patient.
2.5.2 Hertel’s classification
Hertel [36] fundamentally changed the approach by using a binary system based on Lego bricks. He proposed fracture planes instead of fracture fragments. To classify a fracture, possible fracture planes between head and GT, GT and shaft, head and LT, LT and shaft, and finally between GT and LT need to be identified. This results in six options for 2-part fractures, five for 3-part fractures and obviously just one 4-part fracture. In contrast to Neer, Hertel rated any cortical discontinuity as a fracture irrespective of the amount of displacement or angulation. Particular attention has to be paid to seven other parameters, such as the length of the posteromedial metaphyseal head extension, the integrity of the medial hinge with displacement of the shaft in respect to the head, the displacement of the tuberosities, the amount of angular displacement of the head, the occurrence of glenohumeral dislocation, a head impression fracture, a head-split component and the mechanical quality of the bone. 2.5.3 Hertel’s modified classification
Sukthankar et al [37] modified Hertel’s system by replacing numbers with a comprehensive nomenclature. H(ead), G(reater) and L(esser tuberosity) and S(haft) identify possible fracture parts, a fracture plane is represented by a hyphen (-) and represents a cortical disruption between the parts, regardless of displacement and angulation. H-G-L-S
Imaging: • The main purpose of imaging PHF is to determine instability and fragment displacement. • Standardized views are paramount to determine angulation, displacement, and detection of any changes postoperatively. • The axial view displays instability and displacement between shaft and head and should be part of the standard trauma x-ray series. • CT scans should be used for precise fracture analysis and measurement of local bone density. Two- and three- dimensional reconstructions are essential for precise classification and surgical planning. Classification: • Codman’s 4-part model and Neer’s classification (1970) are still the basis for understanding PHFs [31]. • Hertel’s system and the HGLS classification are recommended for more detailed description of the fracture situation. • Other factors like the degree of shaft displacement and angulation/rotation of the head fragment should also be described.
203
Section 3 Fracture management 3.1 Proximal humerus
3
Decision making
Since little high-level evidence exists, there is still much uncertainty about which patients will benefit from non operative treatment, plate fixation, nailing, or arthroplasty [38, 39]. Overall, conflicting results between studies favoring operative intervention and others failing to show much benefit for more displaced and unstable fractures have been described. This demands careful consideration of the patientspecific benefits and risks of operative and nonoperative therapy [22]. Older patients tend to have worse functional outcomes [30]. This trend has been attributed to factors such as fragility, cognitive deficits, rotator cuff injuries, osteoporosis, and poor rehabilitation potential [40]. The indication for surgery in PHFs is usually a relative one. Therefore comorbidities play an important role in deciding whether or not to perform surgery. In cases where deterioration of comorbid medical conditions like renal insufficiency is likely to happen, it is usually better to refrain from surgery even if the fracture type would justify it.
204
General remarks and thoughts: • The severity of comminution in displaced fractures may have a more significant effect on functional outcomes than the choice of treatment; there is also a clear difference in prognosis between 3- and 4-part fractures, but not between 2- and 3-part fractures [5]. In many studies 4-part fractures yielded worse outcomes compared to 2- or 3-part fractures regardless of the treatment chosen [41]. • The functional outcome is difficult to assess, as many variables contribute to a successful patient outcome. Fortunately, functional expectations for older individuals are lower than for younger patients—a less than satisfactory result for a young patient can therefore be completely acceptable for an older person. Even with decreased outcome scores, older patients’ perception of outcome and quality of life can be acceptable [30]. • As evidence supporting routine operative treatment is limited and complication rates are high, decision making should include individual factors such as living situations, comorbidities, and the patient’s attitude towards surgery. With surgeons’ increasing knowledge about appropriate patient selection and limits of specific procedures, results become more predictable. That means, however, that more than one operative method is necessary to address different situations. • If we think about operative fixation, it seems expedient to ask what kind of difference the patient will experience after surgery. There must be some tangible benefit in terms of an increased functional result. This also holds true if an older patient’s functional status benefits from immediate use of the injured extremity by using a cane, for instance (Case 4: Fig 3.1-15). • It is helpful to discuss all relevant aspects among the interdisciplinary team members in addition to the patient and relatives. If the patient seems motivated to make use of an expedited rehabilitation process after surgery and has no contraindications, the patient may receive the same treatment as a younger adult (Case 5: Fig 3.1-16). • With enhanced techniques like the use of fibular strut grafting or allograft bone blocks, better and more reliable results can be achieved both in younger and in geriatric patients (Case 6: Fig 3.1-17) [16, 42]. • As in other areas, outcomes may also be correlated with the level of surgeon experience, the time of surgery and the soft-tissue handling. These potentially important factors are hardly ever reported or investigated in studies and neither is the precise amount of displacement with a standardized CT scan measurement.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
Patient A 90-year-old woman sustained a left pertrochanteric fracture after a fall from standing height and was treated with a proximal femoral nail antirotation (PFNA) plus augmentation. She was living alone and mostly self-reliant with some help from her daughter who lived close by. No signs of dementia and she was cooperative and go-go, ie, fit. Osteoporosis treatment was initiated and the patient has been using a walking cane ever since. Comorbidities • Hypertension • Chronic renal deficiency • Heart failure • Osteoporosis (T-score: spine = -3.6, hip = -3.6). Treatment and outcome Five months later the patient presented with a 2-part surgical neck fracture (HGL-S) after a low-energy trauma at home with medial comminution and varus displacement (Fig 3.1-15a–e). After careful and extensive consultation with the team, the patient and her daughter decided on operative treatment. Stable fixation was achieved with PHILOS augmentation (Fig 3.1-15f–g). Nonanatomical plate position was chosen to maintain the medial support, which was achieved by slight shortening and impaction of the shaft into the head, as well as a slight valgus position of the head fragment. Peri implant augmentation with 0.5 cc of polymethylmethacrylate cement per cannulated screw was used. With this measure, additional fixation was added. Five days after surgery, pain restricted the patient’s mobilization and usage of the affected arm (Fig 3.1-15h); this made it impossible for her to return home as early as possible and to limit care dependency to a minimum. After 12 days the patient was transferred to the internal medicine and rehabilitation department.
a
b
c
CASE 4
Potential benefit of operative fixation in geriatric patients
Six weeks postoperatively a comparison with previous x-rays was not possible, as they were blurred by different projections. Active flexion and abduction was 140°, active rotation in 90°, and abduction 80°. She used a cane to support the right side; she was pain free (i–l) and back home. Other treatment options • Nonoperative treatment: The patient would be unable to use crutches at least for some time because of pain and she would depend on care. It is likely that the fracture would displace more and could lead to a more restricted functional outcome. • Proximal nailing would be a good alternative, as the reduction in valgus and shortening would be beneficial. • Hemiarthroplasty and reverse arthroplasty are not an option for this type of fracture (overtreatment).
d
e
Fig 3.1-15a–l A 90-year-old woman after a low-energy trauma. a–e X-rays showing a 2-part surgical neck fracture with medial comminution and varus displacement. Note the poor bone quality and the shallow head fragment in the computed tomographic scan reconstructions (a–d).
205
Section 3 Fracture management 3.1 Proximal humerus
f
g
CASE 5
i
j
k
l
Fig 3.1-15a–l (cont) A 90-year-old woman after a low-energy trauma. f–g Stable fixation was achieved with PHILOS augmentation. The medial support is maintained by nonanatomical plate position. h X-ray at 5 days postoperative. i–l Six-week postoperative x-rays and clincal image of pain-free patient.
Operative fixation for rapid rehabilitation Patient A 71-year-old woman was living an active social life. She was motivated, cooperative and go-go, ie, fit. The patient sustained a valgusimpacted 4-part fracture (H(c0)-G-L-S) (Fig 3.1-16a–c). The medial head extension was 0 mm, which indicated a high risk of avascular necrosis (Fig 3.1-16d–f). Comorbidities • Arterial hypertension • Varicosis • Polyarthritis • Osteoporosis
206
h
Treatment and outcome The patient opted for surgery. Due to the osteoporotic bone, an anatomomical reduction and stable osteosynthesis with PHILOS augmentation was accomplished (Fig 3.1-16g–i). Immediate active rehabilitation without sling and without relevant postoperative pain led to an excellent active range of motion (ROM) after 3 weeks (Fig 3.1-16j). One year later the fracture healed uneventfully without secondary displacement (Fig 3.1-16k–l). The patient achieved full ROM without pain (Fig 3.1-16m–o).
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
Other treatment options • According to the current authors’ treatment algorithm, displaced 4-part fractures in go-go, ie, fit, patients should be treated operatively. • Nonoperative treatment may lead to consecutive cranial and/or posterior greater tuberosity (GT) displacement with functional impairment in abduction and rotation.
a
d
g
b
• Osteotomies after malunions of the GT usually fail to heal in proper position. With a varus malunion, an anatomical prosthetic solution is also no longer possible. • According to the current authors’ opinion antegrade nailing is not the first choice in 4-part fractures. • Arthroplasty was not an option because a stable and anatomical reconstruction was to be expected.
c
e
h
f
i
Fig 3.1-16a–o A 71-year-old woman with a 4-part fracture. a–c X-rays showing valgus-impacted 4-part fracture. d–f The medial head extension of 0 mm indicates a high risk of avascular necrosis. g–i D ue to the osteoporotic bone, an anatomomical reduction and stable osteosynthesis with PHILOS augmentation was accomplished. Note the rasp ( g) to elevate the head fragment.
207
Section 3 Fracture management 3.1 Proximal humerus
j
m
k
n
l
o
CASE 6
Fig 3.1-16a–o (cont) A 71-year-old woman with a 4-part fracture. j Clinical photograph showing excellent active range of motion (ROM) after 3 weeks due to immediate active rehabilitation without sling and without relevant postoperative pain. k–l One year later the fracture healed uneventfully without secondary displacement. m–o The patient achieved full ROM without pain.
Massive central allograft to ease reduction and prevent secondary displacement of fracture fragments Patient A 65-year-old female patient sustained a low-energy trauma resulting in a 4-part fracture (H-G-L-S). The patient suffered from chronic alcoholism. Although she was not cognitively impaired, her compliance was uncertain. Comorbidities • Osteoporosis with a wide proximal shaft and rarefied cancellous bone Treatment and outcome The head was in varus, the greater tuberosity severely comminuted and ultrashort, ie, without lateral extension that could be fixed directly with the plate. The patient also suffered from osteoporosis
208
with a wide proximal shaft and rarefied cancellous bone, as the computed tomographic scan of the opposite proximal humerus showed (Fig 3.1-17a–c). The head fragment was damaged by the shaft and the patient complained about pain (Fig 3.1-17d). This led the authors to decide on reconstruction with allograft. The allograft resembling a champagne cork (Fig 3.1-17e) locked itself in the shaft and the head fragment and the tuberosities sat on the graft (Fig 3.1-17f–g). The ultrashort greater tuberosity was fixed trans osseously to the graft without additional hardware (Fig 3.1-17h). The intraoperative C-arm follow-ups (Fig 3.1-17i–j) demonstrated the huge allograft supporting the reconstruction (Fig 3.1-17i). Follow-up x-rays were taken after 1 week (Fig 3.1-17k–m).
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
a
b
c
f
d
i
e
j
k
g
h
l
m
Fig 3.1-17a–m A 65-year-old woman with a 4-part fracture after a low-energy trauma. a–c X-rays (a–b) showing the head in varus, the greater tuberosity severely comminuted and ultrashort, ie, without lateral extension that could be fixed directly with the plate. Osteoporosis with a wide proximal shaft and rarefied cancellous bone is visible on the computed tomographic scan of the opposite proximal humerus (c). d–h X-ray showing the head fragment damaged by the shaft (d). Reconstruction performed with allograft that resembles a champagne cork (e) and locked itself in the shaft and the head fragment and the tuberosities sitting on the graft (f–g). The ultrashort greater tuberosity was fixed transosseously to the graft without additional hardware (h). i–j Intraoperative C-arm follow-ups showing the huge allograft supporting the reconstruction (i). k–m Follow-up x-rays at 1 week.
209
Section 3 Fracture management 3.1 Proximal humerus
3.1
Operative or nonoperative?
More or less undisputed indications for surgery even in older patients are: • • • • •
Head-split fracture Fracture dislocation Segmental fracture Open fracture Complicated fracture with additional vascular injury
Most authors also agree that according to the Neer criteria, undisplaced fractures should usually be treated nonoperatively, irrespective of the number of fragments. Moreover, a lack of consistently successful surgical techniques and common complications has resulted in a preference for nonoperative treatment over surgery [6, 43]. Clinical experience has shown that patients regularly benefit from operative fixation even if it is only for the first few weeks after the trauma. The effect of a pain-free extremity on patients, especially on older ones, may be strikingly positive. As mentioned in the introduction of this chapter, it may be difficult to measure this kind of success with traditional scoring systems. Careful decision making and safe procedures, ie, with a limited risk and performed to the best of the surgeon's knowledge, are required, as everything must be done to avoid complications. The worst outcomes usually result from poor open surgery and soft-tissue handling during surgery leading to unstable fracture fixation. It is often observed that the functional status deteriorates after sustaining a PHF. Patients immediately become dependent on help and in many instances require full-time nursing care, at least temporarily. If this could be prevented by operative fixation, which has been demonstrated only by anecdotal evidence, operative fixation would be a good option for some patients. The patient’s living condition also has an impact on making a therapeutic decision. In Table 3.1-1 different aspects and parameters are discussed.
3.1.1 Outcome
At 1-year follow-up of a randomized controlled trial (RCT), Fjalestad et al [44] found no evident difference in functional outcome between operative treatment and nonoperative treatment of displaced PHFs in older patients. Only radiographic scores turned out to be better after operative interventions. Only one RCT exists comparing 3-part PHFs treated nonoperatively versus with a locking plate. Olerud et al [2] investigated 60 patients aged 74 years on average. The results of the study indicate that the locking plate had a positive impact on the functional outcome and health-related quality of life (HRQoL), but at the cost of additional surgery in 30% of the patients. In another RCT, fracture arthroplasty and nonoperative treatment were compared. 55 patients, aged 77 years, ie, 58–92 years, on average, with displaced 4-part fractures were randomly allocated to the two treatment options and monitored for 2 years. The Constant score, ROM, DASH score and pain (visual or verbal analogue scale) did not differ significantly among the patients. The quality of life assessment (EQ-5D) showed significantly better results in favor of surgery [45]. Metaanalyses of 3- and 4-part fractures revealed that patients treated nonoperatively had more pain and a worse ROM than those treated with either fixation or arthroplasty [46]. Van den Broek et al [47] compared antegrade nailing (n = 27) with nonoperative treatment (n = 16). The Constant score was 67.1 in the nailing group and 81.4 in the nonsurgical group. Krettek et al [6] compared two PHF studies dealing with patients after nonoperative and operative treatments respectively, supported by the AO Clinical Investigation and Documentation (AOCID) team. They found more complications (34% versus 28.8%), more revision surgeries (19% versus 7.2%) and a 10% lower Constant score in the surgical group. The extent to which PHFs impact functionality of geriatric patients has not yet been thoroughly investigated. Einsiedel et al [4] described a significant deterioration of walking ability, leading to two or more new falls in 24% of patients with distal radial fractures (DRFs) and 28% of PHF patients in a prospective study of 104 patients.
210
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
Nonoperative treatment
Grey zone
Operative treatment • Patient cannot be managed with ambulatory care because of fracturerelated pain • Crepitation as a sign for instability
Findings in favor of pain
Displacement of the GT
• Nondisplaced • No displacement during follow-up
Displacement shaft
• Nondisplaced and stable • Impacted
Angulation head vs shaft
• < 45° varus/valgus • < 45° anteversion/retroversion
Calcar
• Short GT fragment with lateral comminution and impaction: difficult to address operatively • “Functional 2-part fracture” (with multiple fracture lines of the GT fragment, yet undisplaced), good medial support of the head fragment
• Posterosuperior displacement of GT of > 0.5 cm: GT overlaps the posterior articular surface with loss of external rotation and early glenoidal impingement • Cranialization of GT into the subacromial space • Large GT fragment with high success rate after fixation • > 50% displacement • Unstable medial hinge
• Stable operative fixation is questionable because of medial comminution, allograft is an alternative
• > 45° varus/valgus • > 45° anteversion/retroversion
• Stable surgical fixation is questionable because of medial comminution • Impaction must result in medial stability, otherwise bone grafting
• Calcar can be perfectly reduced
Use of walking aids
• Has to use walking aids
• If the patient cannot be mobilized within the first week, augmented fixation may be considered
Surgical skills and feasability
• Doubts that the surgeon will be able to accomplish it • The worst case is a failed surgery
Concomitant injuries or disabilities
• With cuff arthropathy, nonoperative treatment may be preferred • In case of problems, a reversed arthoplasty is indicated
Low bone quality
• May have an impact on the choice of operative treatment but not on the question of whether operative treatment is indicated
Age, comorbidities, functional status
• No-go or frail patients. They are mostly ≥ 85 years, suffer from three or more comorbidities and geriatric syndromes, and are constantly limited in their daily activities
• Slow-go, intermediate or vulnerable patients may be dependent in one or more IADLs but not ADLs, and suffer from one to two comorbidities but no geriatric syndromes
• Go-go or fit patients are functionally independent in terms of ADLs and IADLs and without serious comorbidities or geriatric syndromes
Compliance, mental status, abuse
• Dementia • In patients with polytoxicomania there is only risk and barely any benefit
• Demanding and cooperative
• Normal or slightly impaired • Highly motivated
Risk of surgery
• High
• Moderate
• Low
• The surgeon is able to do it right because of the size of the fragments, a long cortical extension of the GT fragment and other fracture characteristics like the feasibility of creating intraoperatively intrinsic stability • Compensated cuff arthropathy with good function may be a good nailing indication
• Multilevel injuries
Rehabilitation potential
• Mostly sitting only, needs constant care
• High
Functional expectations
• Low
• High
Financial aspects
No significant difference between operative and nonoperative treatment [44]
Table 3.1-1 Nonoperative versus operative treatment. Factors that may influence decision making in proximal humeral fractures. Items where either direction is possible are in the “grey zone”. Abbreviations: ADLs, activities of daily living; GT, greater tuberosity; IADLs, instrumental activities of daily living.
211
Section 3 Fracture management 3.1 Proximal humerus
3.2
Fixation or arthroplasty?
The question whether stable fixation will be possible is much more important in terms of the predicted outcome than the presumed extent of AVN. Therefore, if stable fixation is possible, there is no need to be concerned about future osteonecrosis. Whether stable fixation is possible may be questionable mainly in 4-part fractures. In critical cases, ie, cases with a low probability of achieving stable fixation, surgeons should be prepared to perform joint replacement. In cases where stable fixation turns out to be impossible and the surgeon does not feel comfortable performing arthroplasty, it may be advisable to get the help of an experienced team to preliminarily fix the fracture with K-wires and perform an early secondary arthroplasty procedure. 3.3
Treatment algorithm
In the authors’ own practices, approximately 30% of all fractures with pronounced comminution and/or dislocation are treated operatively. The majority of all PHFs are minimally or nondisplaced and can therefore be successfully treated without surgery. It should also be taken into consideration that many studies are older and retrospective and focus mainly on objective parameters like ROM or x-rays. Patients are getting older nowadays and attitudes and demands may change. Fragile patients have limited reserves of physical strength and even short-term functional decline may be difficult to compensate. Some patients may therefore benefit from operative fixation to restore functionality earlier. Obviously, one type of treatment does not fit all pathologies and patient profiles. Most surgical methods can be applied to a variety of fracture types, but they each have strengths and weaknesses. Avoiding complications is one of the main goals, so surgeons should be able to choose between modalities servicing their actual need, which also includes arthroplasty. The following questions need to be answered in daily practice: • Are the fracture fragments displaced enough to require surgery? • Is the patient a good candidate for surgery ie, mentally fit and demanding? • Is the surgical risk within the normal range? • Is it unlikely to produce a surgical complication? The type of treatment will be decided based on the criteria listed in Table 3.1-2.
212
3.4
Summary
Decision making: • Though improving, evidence is not yet sufficient to allow for robust treatment recommendations. Surgeons’ skills and preferences play an important role. Most current treatment recommendations are based on expert opinion and low-powered studies. • Treatment with locked plates and antegrade nails has failed to demonstrate better results than nonoperative treatment and is fraught with a complication rate of ~ 30%. • Selection of patients suitable for a particular treatment seems to be of paramount importance. Criteria for a successful outcome have not yet been fully identified. • In patients older than 60 years, nonoperative and operative protocols, including arthroplasty, yield similar functional results despite better x-rays after operative fixation. Operative indications must therefore be well justified individually. • Obvious operative indications are fracture dislocations, head-split fractures, open fractures, pathological fractures, and segmental fractures. • For patients unfit for surgery or with important risk factors indicating deterioration after operative, surgical procedures should only be suggested with great caution and in accordance with the whole orthogeriatric team. • Nonoperative treatment is indicated in simple and nondisplaced PHFs. The same approach may apply in selected cases of more complex injuries and if patients are unfit for surgery. • Arthroplasty should only be chosen if all other options will presumably fail.
Grade
Explanation of activity
0
Fully active, able to carry on all predisease performance without restriction
1
Restricted in physically strenuous activity but ambulatory and able to carry out work of a light or sedentary nature, eg, light house work, office work
2
Ambulatory and capable of all self-care but unable to carry out any work activities. Up and about more than 50% of waking hours
3
Capable of only limited self-care, confined to bed or chair more than 50% of waking hours
4
Completely disabled. Cannot carry on any self-care. Totally confined to bed or chair
5
Dead
Table 3.1-2 World Health Organization performance status.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
4
Therapeutic options
Apart from nonoperative treatment, several operative options are at the surgeon’s disposal, but there are no evidencebased recommendations to specifically guide selection. General remarks and thoughts: • Soft-tissue handling is critical for a successful outcome, but it has not been studied in detail. It is the factor often not considered in any study and may not be comparable even if the type of fracture, approach, and implant are the same. • Open fixation must result in a stable construct which allows for immediate postoperative physiotherapy. If this is not achievable, a prosthetic replacement should be used. Sometimes it is necessary to switch strategies during an operation. • Angular stable implants alone do not solve the problem. Anatomical reduction with either cortical contact or void filling must also be provided in order to prevent secondary displacement of parts with subsequent cut-through or cut-out of screws. • It seems important to mention again that nonoperative treatment is more desirable than a poorly performed operative procedure regardless of the method of fixation [13]. The surgeon must know the chosen method very well. • At the moment, discussion about anatomical hemiarthroplasty and reverse arthroplasty is open; there are no clear guidelines for the orthopedic trauma surgeon to treat these disabling fractures in geriatric patients.
4.1
Nonoperative treatment
The vast majority of PHF are successfully treated nonoperatively. The rotator cuff, remnants of periosteum and capsular tissue (so-called ligamentotaxis effect) often provide enough intrinsic stability to resist further displacement of fracture fragments. Minimal tuberosity displacement combined with controlled shaft impaction reduces the risk of nonunion. 4.1.1 Pain treatment
Initial pain control after the injury includes a combination of oral medications, topical modalities, and sling immobilization. Providing adequate pain control over the first couple of days after injury is of paramount importance for functional recovery. Regional nerve blocks like the interscalene block, ie, Winnie block, the supraclavicular perivascular (subclavian perivascular) block or the suprascapular nerve block should be considered. Exercises should never be painful, otherwise a complex regional pain syndrome may result. 4.1.2 Fracture reduction
Attempts to reduce a PHF have little, if any, effect on rates of malalignment or functional outcomes. Given the potential risk for soft-tissue and plexus injuries caused by manipulation, surgeons should critically reconsider the indication for fracture reduction (Case 7: Fig 3.1-18) [48]. Moreover, the typical varus malalignment may develop after several weeks (Case 8: Fig 3.1-19) and is not predictive of the outcome [49].
213
Section 3 Fracture management
CASE 7
3.1 Proximal humerus
Spontaneous realignment with nonoperative treatment Patient Nonoperative treatment of a proximal humeral fracture (PHF) in an 87-year-old woman.
Treatment and outcome Five years before PHF the patient sustained a displaced, intraarticular right distal radial fracture (Fig 3.1-18a) that healed after nonoperative management with acceptable malalignment (Fig 3.1-18b–c). Initially, the 2-part PHF was minimally displaced (Fig 3.1-18d–e). The patient was treated with a sling. After 1 week pronounced displacement could be seen in a true AP view (Fig 3.1-18f). The fracture was still markedly displaced 2 and 4 weeks later (Fig 3.1-18g–h). After 6 weeks the situation improved (Fig 3.1-18i) and after 10 weeks the fracture progressed to healing in a good position (Fig 3.1-18j–k).
a
f
b
g
c
h
d
i
j
e
k
Fig 3.1-18a–k An 81-year-old woman with a 2-part fracture. a–e X-rays showing a displaced, intraarticular right distal radial fracture (a) that healed with acceptable malalignment after nonoperative treatment (b–c). Initial displacement of the 2-part proximal humeral fracture occurred (d–e). f–k True AP view (f) at 1 week showing pronounced displacement. X-rays showing the still markedly displaced fracture at 2 and 4 weeks ( g–h) but improved situation at 6 weeks (i) and fracture progressing to healing in good position at 10 weeks ( j–k).
214
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
Patient An 81-year-old woman in excellent condition sustained a 2-part fracture. Comorbidities • No relevant comorbidities but obvious radiological signs of severe osteoporosis. Treatment and outcome The 2-part fracture (Fig 3.1-19a–b) showed obvious radiological signs of osteoporosis (25-hydroxyvitamin D3: 11.2 ng/mL [28 nmol/L]; PTH: 56 μg/L; T-score: lumbar spine = -2.0; proximal femur = -2.2). The patient was administered 3 mg IV of ibandronate every 3 months, 1,000 mg of calcium, 400 IU of vitamin D. Other wise, the patient was in excellent condition without relevant comorbidities.
a
f
b
g
CASE 8
Typical course of nonoperative treatment
Nonoperative treatment was chosen. The patient’s arm was put in a sling, adequate pain treatment was administered (Fig 3.1-19c–d), and the arm was mobilized as soon as possible. After 4 weeks the alignment was good and the patient had little pain (Fig 3.1-19e). After 9 weeks the fracture was well aligned and the function improved (Fig 3.1-19f–g). Five months after the injury the typical varus malalignment by continuous pull of the supraspinatus muscle occurred (Fig 3.1-19h–i). The function was excellent and the patient was satisfied with the outcome (Fig 3.1-19j–m). At the 5-year follow-up, the patient broke her right proximal femur which was treated with a proximal femoral nail antirotation augmentation (Fig 3.1-19n–o). The T-score of the opposite hip was -3.4 and of the lumbar spine was -2.3. A computed tomographic scan of her right shoulder revealed a stable nonunion of the proximal humeral fracture without any complaints and near to normal function (Fig 3.1-19p–q).
c
h
d
i
e Fig 3.1-19a–q An 81-yearold woman with a 2-part fracture. a–b X-rays showing a 2-part fracture with obvious radiological signs of osteoporosis. c–e X-rays after nonoperative treatment, ie, arm in sling (c–d), showing good alignment (e). f–g Well-aligned fracture after 9 weeks. h–i X-rays 5 months post injury showing typical varus malalignment.
215
Section 3 Fracture management 3.1 Proximal humerus
j
n
k
l
o
p
q
Fig 3.1-19a–q (cont) An 81-year-old woman with a 2-part fracture. j–m C linical photographs showing excellent function. n–q F ive-year follow-up x-rays showing the patient’s right proximal femur fixed with a proximal femoral nail antirotation augmentation (n–o). Computed tomographic scans of the right shoulder revealing a stable nonunion of the proximal humeral fracture ( p–q).
4.1.3 Immobilization
4.1.4 Outcome
Generally, the treatment strategy should be as functional as possible, allowing patients to use their injured shoulder as much as pain allows (Table 3.1-1). Prolonged immobilization with a Gilchrist type of sling has not been proven to be effective. The authors encourage patients to use their upper extremity as much as possible while performing daily activities. With clinical and x-ray follow-ups, those fractures that are more unstable than initially anticipated can be identified. If necessary, the treatment strategy may be changed eventually (Case 9: Fig 3.1-20).
Radiologically, nonoperative treatment usually results in some malalignment. In 2-part fractures this typically leads to a varus position of the proximal fragment with a subacromial position of the GT. In contrast to younger patients and to patients after operative treatment, this does not interfere with a reasonable or even good function of the shoulder in older patients [49].
A slab splint from the forearm to the shoulder or Desault’s bandage may cause iatrogenic problems like skin abrasions, swelling, or stiffness. The predicted risk of delayed and/or nonunion is 7% with nonoperative management. The risk for nonunion in smokers is 5.5 times higher than in nonsmokers.
216
m
Retrospective studies have shown a near functional normality in 80% of geriatric patients with only minor restrictions in strength and ROM to vigorous activities especially if the fracture is only minimally displaced [50, 51]. Most patients should be able to perform activities of daily living (ADLs) (Case 10: Fig 3.1-21). There is robust evidence that nonoperative management of PHFs is safe and effective, mainly in AO/OTA type A and B fractures [5].
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Secondary displacement with change of treatment modality Treatment and outcome There existed a trauma series of the fracture including an axial view without relevant displacement. The patient expressly said that she did not want operative treatment so ambulatory nonoperative treatment was started (Fig 3.1-20a–c).
Patient An 83-year-old woman with severe osteoporosis and a 2-part fracture after a fall at home. She was go-go, ie, fit, had no relevant comorbid conditions, and took only one medication for brady arrhythmia.
CASE 9
Franz Kralinger, Michael Blauth
The 1-week follow-up showed complete dislocation and the patient had distinct pain despite oral pain medication (Fig 3.1-20d–f). Operative fixation was therefore recommended. One day later, operative fixation with 9.5 mm/160 mm intra medullary nail, statically locked, 3 head screws, one screw in screw (Fig 3.1-20g–h). Three months later there was no change in position, bone healing, and excellent function (Fig 3.1-20i–k). Two years later she presented again in the emergency department with a knee distortion after lifting a clothes basket.
a
b
d
i
c
e
j
f
k
g
h
Fig 3.1-20a–k An 83-year-old woman after a fall at home. a–c A trauma series of the 2-part fracture including an axial view without relevant displacement. d–f Follow-up after 1 week showing complete dislocation. g–h X -rays showing fracture treated operatively with 9.5 mm/160 mm intra medullary (IM) nail, three locked screws in proximal end of IM nail and standard distal interlocking screws. i–k X -rays after 3 months showing no change in position, bone healing, and excellent function.
217
Section 3 Fracture management
CASE 10
3.1 Proximal humerus
Adequate functional result despite massive malalignment in a very old patient Patient A 96-year-old female patient with a 3-part proximal humeral fracture (HL-G-S) after a fall at home. Treatment and outcome Nonoperative treatment was chosen because of the patient’s age. The greater tuberosity (GT) was pulled posteriorly and superiorly (Fig 3.1-21a–e).
Discussion Progression into varus malalignment within the first 6–8 weeks is frequent in patients treated nonoperatively. Large series show little functional impairment in 2-part surgical neck fracture. In contrast, posttraumatic malunion of the GT is problematic to treat. Reverse arthroplasty is the preferred choice in this age group, yet the amount of external rotation depends on the displacement of the GT limiting range of motion in the “hinged door” mechanism of the reversed polarity.
Six months later the head fragment was displaced more into varus and anteversion (Fig 3.1-21f–h). Some anterior part of the GT was still attached to the head fragment which allowed some supraspinatus muscle functions. Posterosuperior displacement of the GT occurred. The pain was tolerable and the patient could flex the arm 150°, reach her mouth and the back of her head; she could also reach her buttocks.
a
b
f
g
c
d
e
h
Fig 3.1-21a–h A 96-year-old woman after a fall. a–e X-rays showing the greater tuberosity (GT) pulled to posterior and superior—note the intact supraspinatus muscle (SSP) (arrows). f–h Six months later the head fragment was displaced more into varus and anteversion. Some anterior part of the GT was still attached to the head fragment which allowed some SSP functions. Posterosuperior displacement of the GT.
218
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
4.2
Locking plate
After the introduction of the locking plate concept in the early 1990s, it rapidly became the most frequently used technique for fixation of PHFs. This still holds true today despite an unacceptably high rate of reported complications. Many patients, however, achieve good to excellent results. Factors predicting failure are understood better nowadays and can be avoided more easily. Regardless of the approach, soft-tissue management seems to be at least as important as biomechanical issues. No data exists on the impact of soft-tissue handling on the outcome, since it is very difficult to control this in studies. Topic 4.2 focuses on lessons that have been learned over the past years. For a systematic description of the technique, see Acklin et al [52] and Plecko et al [53]. 4.2.1 Approaches
Deltoid-split approach for minimally invasive plate osteosynthesis (MIPO): • The use of this minimally invasive approach should not result in malreduction, specifically of the varus and valgus positions of the head fragment. • The surgeon should be familiar with MIPO techniques such as indirect reduction and the use of joysticks as the fracture zone cannot be visualized directly. • It offers complete access to the entire GT, especially when parts of it are posteriorly displaced, and to the plating area. • On the other hand, there is an increased risk of limited access to the head fragment, and the axillary nerve is in danger. Options to extend the approach are limited. • In a comparative study with the deltopectoral approach, the use of the deltoid-split approach resulted in a lower Constant score due to poorer ROM. According to a retro spective study there is no difference between the two approaches including electrophysiological investigations [54, 55].
Deltopectoral approach (Approach 1: Fig 3.1-22): • Although the skin incision is longer in this approach than in the deltoid-split approach, the procedure should be as minimally invasive as possible, which can be achieved by preserving periosteal bridges, using sutures to manipulate the fracture fragments, and using the plate to indirectly reduce the fracture. For excellent exposure, no deeper structures need to be dissected. • Manipulation of tuberosities should only be performed by sutures through the tendon of the rotator cuff. They should never be pulled directly to avoid further fragmentation and iatrogenic periosteal detachment. • The affected arm should be positioned on a Mayo side stand in abduction. By internal rotation and abduction, the deltoid muscle can be retracted and the GT exposed. • Rounded retractors are inserted underneath the deltoid muscle. • A longitudinal split of the SSP tendon does not cause important damage and can be closed easily. The quality of the GT reduction can be clinically assessed with the SSP split and is also easy to extend. • The so-called tendon-free rotator interval between SSC and SSP muscles should not be dissected. The whole complex together with smaller anterior fracture fragments act like a hood that can be pulled over the humeral head at the end of the surgery. • This approach is used for revision surgery as well.
219
Section 3 Fracture management
APPROACH 1
3.1 Proximal humerus
a
b
c
d
e
f
Fig 3.1-22a–f Deltopectoral approach. a The skin incision runs along the humerus, starting lateral to the palpable tip of the coracoid. b–c T he cephalic vein is retracted laterally and the subdeldoital space is opened by exposing the clavipectoral fascia. The lateral tendon border of the conjoined tendon and the coracoacromial ligament are identified (not shown in the pictures). By simple abduction of the arm on a Mayo side stand the tension of the deltoid is relieved and the exposure can be nicely extended laterally (c). d No soft-tissue structures are dissected or stripped including those already injured. e Stay sutures through the cuff are placed from anterior to posterior. The anterior suture allows manipulation of the head to place the next one until the posterior cuff is reached. The tuberosities are never manipulated with sharp instruments. f The authors’ preferred method is first to fix the plate to the shaft and then to manipulate and temporarily fix the fracture with pins without interfering with the plate position. Two pins, one just below the bony edge of the greater tuberosity and one just behind the bicipital groove, are used for guidance of the final plate position.
4.2.2 Biceps tendon
Sutures or cerclages crossing the bicipital groove with the tendon underneath act like a tenodesis (Fig 3.1-23, Fig 3.1-24, Case 11: Fig 3.1-25). Therefore, whenever this approach is necessary, surgeons are advised to tentomize the biceps tendon, perform a tenodesis by suturing the distal stump to the insertion of the pectoralis muscle and resect the proximal part of the tendon. 4.2.3 Typical complications
Screw cut-through is typically caused by subsidence of the head fragment over the locked screws. This is possible because of some comminution at the fracture zone and the big central void that characterizes the proximal humerus in the elderly. Subsidence may be combined with varus displacement of the head fragment and usually takes place in the first several weeks after the surgery; it should not be confused with AVN. Typically, the screw tips perforate the head fragment and damage the glenoid fossa. Changing the screws immediately may prevent serious damage to the joint (Case 11: Fig 3.1-25a–j).
220
Screw cut-out goes hand in hand with a loss of reduction. Driven by the pull of the rotator cuff, the humeral head “moves” into a varus position. Cut-through and cut-out often occur at the same time and may lead to irreversible destruction of the glenoid if action is taken belatedly. Revision is challenging and often ends with reverse arthroplasty as a last resort. Boileau et al [56] described the “unhappy triad after locking plate” fixation as a combination of: • Humeral head necrosis • Loss of reduction with posterior migration of the GT and posterosuperior cuff insufficiency • Glenoid erosion and destruction because of screw penetration
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
a
b
c
Fig 3.1-23a–c a–b Whenever sutures cross the sulcus of the biceps, the physiological gliding of the biceps in the groove is made impossible. Adhesion between the interval tissue and the biceps tendon, which is no longer gliding, generates pain and loss of external rotation. Therefore, in these cases a tenodesis of the long head of the biceps (LHB) tendon is recommended. The authors’ preferred technique is to suture the tendon to the upper edge of the pectoralis major tendon. The biceps tendon is then cut proximally and the intraarticular portion is harvested through the fracture plane between the tuberosities. c Note the plane of the intertuberosity fracture, running 8–10 mm behind the sulcus, as the sulcus provides the best bone quality in this area. Frequently the supraspinatus muscle (SSP) tendon is already split minimally at the end of the fracture plane. If needed, eg, in the case of a head-split fracture, when resection of the intraarticular portion of the LHB is difficult, the split of the SSP tendon can be prolonged in line of the fibers and the whole posterior part of the cuff can be pushed back like a curtain providing excellent exposure to the head fragment.
a
b
c
d
Fig 3.1-24a–d The long head of the biceps tendon is first sutured to the upper border of pectoralis major tendon and then cut above. The intraarticuar portion can be harvested without compromising the interval through the existing fracture between the tuberosities. If needed, the supraspinatus tendon can be split along the fibers to extend the needed exposure. Likewise, this method to extend the intraarticular exposure can be used in head-split reconstruction.
221
Section 3 Fracture management
CASE 11
3.1 Proximal humerus
Typical “mechanical” complication due to varisation and subsiding of the head fragment Patient An 83-year-old patient, living in a nursing home, suffered a fall from standing height. Comorbidities • Hypertension • Aortic valve replacement • Mitral valve insufficiency • Atrial fibrillation • Left bundle branch block • Coronary heart disease after bypass surgery • Multiple vertebral fractures • Mastectomy after breast cancer 26 years ago
X-rays after 11 days showed severe displacement and the patient had great pain (Fig 3.1-25f–g). The geriatric evaluation rated the patient as slow-go, ie, unfit, and pain treatment was adapted. Mobilization and pain control deteriorated and after another 8 days of inpatient treatment the decision in favor of surgery was made. Intraoperative x-rays showed good alignment and plate fixation. Medial support reconstruction was questionable (Fig 3.1-25h–j). Within 5 weeks the head fragment subsided and screw penetration occurred. Revision surgery with shorter screws was performed. Arthroscopy demonstrated erosion of the glenoid surface (Fig 3.1-25k–l).
Treatement and outcome The patient sustained a 3-part fracture (HG-L-S) with sufficient alignment for nonoperative treatment (Fig 3.1-25a–b). The computed tomographic scan confirmed the diagnosis and classification (Fig 3.1-25c–e).
Within 1 year the head collapsed (Fig 3.1-25m–n). The patient refused revision surgery, as she was pain free. The function was poor, though. Within the same year she suffered a pertrochanteric fracture that was treated with a long proximal femoral nail antirotation (Fig 3.1-25o–q). Four years later she suffered a periimplant humeral shaft fracture that was treated with a long locking compression plate (Fig 3.1-25r–s). Discussion Retrospectively, an important reason for failure was the lack of medial support, a lesson learned in the past. A well-fixed straight antegrade nail with impaction might have resulted in sufficient stability for healing. The further course was characterized by an overall poor result, but pain was still tolerable. Also in this age group, secondary fracture prophylaxis was required.
a
c
222
b
d
e
Fig 3.1-25a–s An 83-year-old patient after a fall from standing height. a–b X-rays showing a 3-part fracture with sufficient alignment for nonoperative treatment. c–e Computed tomographic scan of the fracture.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
f
k
o
g
h
i
j
l
m
p
q
n
r
s
Fig 3.1-25a–s (cont) An 83-year-old patient after a fall from standing height. f–g X-rays after 11 days showing severe displacement. h–j Postoperative x-rays after surgery showing good alignment and plate fixation. k–l Within 5 weeks the head fragment subsided and screw penetration occurred. Revision surgery with shorter screws was performed. Arthroscopy demonstrating erosion of the glenoid surface (l). m–n Within 1 year the head collapsed. o–q Pertrochanteric fracture treated with a long proximal femoral nail antirotation. r–s Periimplant humeral shaft fracture treated with a long locking compression plate.
223
Section 3 Fracture management 3.1 Proximal humerus
Boileau et al [56] also stated that a failure after a locking plate fixation cannot be reversed. A revision with hemiarthro plasty is not possible because of glenoid erosion and GT migration. An anatomical total shoulder arthroplasty cannot be performed for the same reasons, specifically because of the posterosuperior cuff insufficiency. Unfortunately, reverse shoulder arthroplasty (RSA) will yield poor functional results because of stiffness and absence of the external rotator muscles. There are several ways to avoid these complications: • Careful evaluation of the thickness of the head fragment in the preoperative CT scan. If it is below 8–10 mm, the risk increases and additional measures like augmentation should be considered, especially in osteoporotic bone (Case 3: Fig 3.1-14). • Careful “percussion” drilling and measuring screw lengths without perforating the articular surface. Perforation of the subchondral bone increases the risk drastically. A blunt probe for length measurement should be used. • Filling the void after reduction and supporting the humeral head are other options to prevent fragments from being displaced. Techniques that used allograft bone or fibular strut graft [57, 58] have turned out to be more successful even in unfavorable conditions [16, 42]. Ideally, massive bone blocks from femoral heads are available from a local bone bank (Case 12: Fig 3.1-26). This applies specifically to those cases with a shallow head fragment (which is particularly pronounced in osteoporotic cases), comminution zones, and/or a short or multifragmentary GT fragment that would not be able to take any screws but can only be fixed with sutures. • Enhancing the anchorage of the screws in the reduced cancellous subchondral bone is another approach that is currently under evaluation. Injecting small amounts of PMMA cement through cannulated and perforated screws leads biomechanically to significantly more cycles to cut out. This only works in osteoporotic bone structure. • Medial hinge (calcar) reconstruction either by meticulous reduction or in case of comminution by shortening to achieve mechanical stability. Indirect techniques like employing a bone rasp to reduce the head fragment through the fracture can be used. When the plate is already in situ, reduction can be secured by using a preliminary pin through the plate. The quality of medial calcar restoration should be carefully checked by image intensification.
224
• Controlled impaction of the head fragment onto the humeral shaft or vice versa in case of medial comminution and highly osteoporotic bone quality. Impaction may contribute to the overall stiffness and has to be done correctly, ie, no varus malalignment must result. • Accepting a nonanatomical position of the PHILOS plate after shortening. When approximating the humeral shaft to the plate, the achieved medial support will be lost (Case 13: Fig 3.1-27). If the head is impacted, the plate cannot be applied in an anatomical position (Case 13: Fig 3.1-27). Frequently the shaft is shifted laterally by the first screw and pulled out of the medial support, which is to be avoided. • Under no circumstances should a residual varus position of the head fragment be accepted. A slight overcorrection into valgus may be beneficial and is preferred. • The GT functions as a capstone that completes the corpus and offers intrinsic stability at least in cases with reasonable bone quality. The implant must then only keep the position, as loads are transferred by the bone. • To support the medial corner and to prevent varus mal alignment, calcar screws have proven to be effective [59]. • In situations with lateral comminution of the GT and/or a short GT fragment that is fractured close to the tendon, insertion the standard plate position does not address the superior part of the GT any more. Additional suture fixation of the GT through the SSP tendon-bone-interface to the prepared holes in the plate may be particularly helpful. In cases with a big central void it is of paramount importance that enough bony substance underneath the GT and the head fragment is provided to restore mechanical stability and promote healing. A massive allograft works well in these cases. • Isolated screws should not be used for fixation of the lesser tuberosity in this population. The screws often loosen and may cause problems culminating in revision surgery.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Central massive allogenic bone block to prevent secondary loss of reduction Patient An 89-year-old woman sustained a 2-part proximal humeral fracture (PHF) with additional metaphyseal fragments after a fall in her nursing home (Fig 3.1-26a–b). Her World Health Organization performance status was 4, Parker Mobility Score 3, and she used a cane; CAM 0, Lachs Score 10, local bone mineral density 70.9 mg/cm3. She had sustained an L1 fracture many years ago (Fig 3.1-26c). Comorbidities • Parkinson´s disease • Epilepsy
l
A computed tomographic scan after 6 months showed integration of the graft without loss of reduction and also a reasonable function (Fig 3.1-26j–l). Discussion Nonoperative management must be discussed in this case. If surgery is chosen, additional measures apart from fixation are necessary, as otherwise failure of any fixation is likely.
a
g
Treatment and outcome The fracture was operatively reconstructed with a massive allogenic bone block from the local bone bank (Fig 3.1-26d–f). Three months later the fracture healed uneventfully (Fig 3.1-26g–i).
CASE 12
Franz Kralinger, Michael Blauth
b
h
c
e
d
i
j
m
k
f
Fig 3.1-26a–m An 89-year-old woman after a fall. a–b X-rays showing the 2-part proximal humeral fracture with additional metaphyseal fragments. c X-ray of L1 fracture taken 2 years prior to the proximal humeral fracture. d–f Operative reconstruction with massive allograft bone block from the local bone bank. g–i X -rays after 3 months showing uneventful healing of the fracture. j–k Computed tomographic scan at 6 months showing integration of the graft without loss of reduction. l–m F ollow-up range of motion examination.
225
Section 3 Fracture management
CASE 13
3.1 Proximal humerus
Importance of calcar reconstruction Patient A 79-year-old female patient who experienced a fall at home. The patient sustained a displaced 3-part proximal humeral fracture with a head-split component (Fig 3.1-27a–d). Comorbidities • Hypertension • Hyperlipidemia • Osteoporosis
a
h
m
b
c
i
d
j
n
Treatment and outcome The fracture was reduced and preliminarily fixed with K-wires (Fig 3.1-27e–g) and the calcar was reconstructed (Fig 3.1-27h–l). Moving the shaft to the plate would lead to uncontrolled impaction of the shaft into the head of the humerus with loss of reduction (Fig 3.1-27h–i). After 4 weeks (Fig 3.1-27m–n) and 1 year (Fig 3.1-27 o–q) there was no change in position. Constant score right/left was 78/77. Range of motion and power were almost equal on both sides.
e
k
o
f
g
l
p
q
Fig 3.1-27a–q A 79-year-old woman after a fall. a–d X-rays showing a displaced 3-part proximal humeral fracture with a head-split component. e–g X-rays showing reduced fracture and preliminarily fixed with K-wires. h–l X-rays showing reconstructed calcar. Note the distance of the plate from the humeral shaft (arrow). Moving the shaft to the plate would lead to uncontrolled impaction of the shaft into the head of the humerus with loss of reduction (h–i). Postoperative follow-up x-rays ( j–l). m–q No change in position after 4 weeks (m–n) and 1 year (o–q).
226
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
4.2.4 Aftercare
4.3
Postoperative physiotherapy and self-directed exercises play a crucial role in achieving a good outcome. The stability of the fixation should always allow for immediate and active treatment.
Antegrade humeral nailing has gone through an evolution with regard to nail and screw designs. With improved options of the latest nails, this type of osteosynthesis has become an alternative to plate fixation.
Patients should never experience pronounced pain, ie, a differentiated pain regimen including nerve blocks is vital. Passive mobilization is allowed from day 1, active motion as soon as pain subsides. Using a sling is not mandatory but many patients feel more comfortable with a sling at least in the first couple of days.
A straight nail design seems to be superior to a curvilinear one for many reasons [63]. With a modular concept, implant configuration can be adapted to fracture morphology.
We encourage patients to stretch the elbow and use the hand of the affected arm as soon as possible for ADLs. 4.2.5 Outcome
Comparing results from different studies is challenging, because the amount of displacement and angulation is rarely quantified and the term unstable is not clearly defined and is difficult to measure preoperatively. Complications are hardly ever monocausal. Krappinger et al [28] reported that age, local BMD, anatomical reduction, and restoration of the medial cortical support to be important predictors for failure. The probability of mechanical failure increased significantly when at least three of the following risk factors were present: higher age, lower local BMD, less than anatomical reduction, no medial cortical support. A thorough and prospective analysis of 150 patients with unstable PHFs failed to show any of the suspected parameters to be significant including surgical characteristics, smoking, alcohol consumption, fracture type, medical comorbidities, and delay of surgery [1]. Loss of reduction and secondary screw loosening with perforation or both were the most frequent mechanical complications.
Antegrade nailing
Nail insertion via a split of the SSP tendon is more widely accepted in older than in younger patients. Antegrade nailing is most often indicated in displaced 2-part PHFs. 4.3.1 Advantages
• It is a minimally invasive surgical procedure. • Biomechanically, an IM nail comes closer to the calcar than a plate and may support this important structure more effectively (Case 14: Fig 3.1-28). This can be further enhanced by an ascending calcar screw and the so-called screw-in-screw concept to address the posteromedial area of the humeral head [64]. • Some nails offer a modular system with angular stable proximal and distal locking options. • Ipsilateral segmental fractures of the proximal humerus and the humeral diaphysis can be stabilized by using a long version of the nail. 4.3.2 Disadvantages
• These demanding operative techniques require hands-on training so as to avoid errors and complications. • Subacromial scarring may cause painful limited ROM. • Despite improved locking options, loss of reduction and varus malalignment still occur (Case 15: Fig 3.1-29). For a systematic description of the technique, see Hessmann et al [65].
Spross et al [60] reported 294 patients, aged 72.9 years on average, with a complication rate of 28.2% and 24.5% required revision surgeries. Screw cut-out was the most frequent reason due to secondary displacement. Smokers with more than 20 pack years had significantly more complications. Other studies reported similar rates [3, 61]. Clearly, patients with complications have worse functional outcomes [1, 62].
227
Section 3 Fracture management
CASE 14
3.1 Proximal humerus
Straight intramedullary nail to address unstable 2-part fracture Patient A 91-year-old female patient fell in her nursing home in the middle of the night. She suffered a massive hematoma, but her neurovascular status was intact.
Treatment and outcome The patient sustained a displaced 2-part proximal humeral fracture with metaphyseal comminution (Fig 3.1-28a). She was under general anesthesia when closed reduction and fixation with an 8 mm intramedullary (IM) nail was performed (Fig 3.1-28b–e). Postoperative AP and lateral x-rays showed good alignment (Fig 3.1-28f–g).
Comorbidities • Urinary tract infection with Escherichia coli • Atrial fibrillation after myocardial infarction 5 years ago • Chronic obstructive pulmonary disease
After 6 months the patient sustained two minor strokes. The fracture healed in anatomical position. She used a wheelchair, did not report any pain and had equal range of motion on both sides (Fig 3.1-28h–i). Discussion This case shows the power of IM nailing. With IM nailing surgeons should strive for a medial support. In this case, it was decided not to open up the fracture and to take some risk in terms of stability.
a
b
c
h
228
d
i
e
f
g
Fig 3.1-28a–i A 91-year-old woman after a fall. a A displaced 2-part proximal humeral fracture with metaphyseal comminution. b General anesthesia and positioning of the arm with an arm holder. c–e Closed reduction and fixation with an 8 mm intramedullary nail. f–g Postoperative result with good alignment in AP and lateral views. h–i X-rays showing fracture healed in anatomical position.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Typical course after intramedullary nailing of a 2-part proximal humeral fracture Patient An 81-year-old, clearly go-go, ie, fit, male patient sustained a displaced 2-part proximal humeral fracture 4 days after a fall (Fig 3.1-29a–b). Two years previously, he had sustained a type A fracture of L1 with no treatment of the underlying osteoporosis (Fig 3.1-29c).
Treatment and outcome The fracture was anatomically reduced and fixed including a calcar screw (Fig 3.1-29d–e). One week (Fig 3.1-29f), 3 weeks (Fig 3.1-29g), 6 weeks (Fig 3.1-29h), and 8 months (Fig 3.1-29i–k, Fig 3.1-29j in internal rotation) after surgery there was limited loss of reduction into varus and osseous healing. The patient experienced residual pain; flexion/abduction was 100°.
Comorbidities • Myocardial infaction • Type 2 diabetes • Chronic renal insufficiency
a
f
b
c
g
h
CASE 15
Franz Kralinger, Michael Blauth
d
i
e
j
k
Fig 3.1-29a–k An 81-year-old man after a fall. a–c X-rays showing a displaced 2-part proximal humeral fracture. Note the amount of displacement in the axial view. d–e Anatomical reduction and fixation including a calcar screw. f–k X-rays 1 week (f), 3 weeks ( g), 6 weeks (h), and 8 months (i–k, j in internal rotation) after surgery showing limited loss of reduction into varus and osseous healing.
229
Section 3 Fracture management 3.1 Proximal humerus
4.3.3 Typical complications
Suboptimal initial reduction:
Iatrogenic rotator cuff injury: • This is the main reason why many surgeons are reluctant to use IM antegrade fixation techniques in younger adults. Since the rotator cuff already shows some signs of degeneration, this concern is of less importance in older adult patients. • Nails with a straight design are inserted through the muscular part of the SSP and the superior part of the humeral head, thus potentially avoiding the delicate tendinous part at the footprint of the tendon. • Euler et al [16] analyzed CT scans of bilateral proximal humeri in 200 patients with an average age of 45.1 years (SD 19.6; 18–97) without humeral fractures. They defined the entry point of the nail and the region of interest, ie, the biggest entry hole that would not encroach on the insertion of the SSP tendon. This showed that 38.5% of the humeral heads had to be categorized as critical types due to morphology in which the predicted offset of the entry point would encroach on the insertion of the SSP tendon that might damage the tendon and reduce the stability of fixation. The authors recommended studying the preoperative x-rays accordingly and to choose another treatment option in critical cases (Case 16: Fig 3.1-30). Subacromial impingement:
• Preexisting osteoarthritis of the shoulder, shoulder stiffness, and inability to recline the arm may be serious obstacles to reaching the optimal entry portal. • Manipulation of fragments may be difficult. Only minimally invasive techniques such as K-wires and sutures (eg, joystick, stay suture) should be applied. • In addition to the correct entry point, the direction of the bony canal in the humeral head is extremely important; it must be parallel to the humeral shaft both horizontally and vertically and cannot be changed. The surgeon needs to get it right the first time, otherwise the head fragment may break (Case 17: Fig 3.1-31). Loss of reduction: • The central entry point offers an additional “fifth” anchor point in the humeral head. Prior to the operation the surgeon should check carefully if the bone quality around the entry portal is sufficient. A wrongly chosen entry portal cannot be changed. If fixation will not be stable, the surgeon must be able to switch to a plate or an arthroplasty. • Locking screws in the humeral head should aim towards areas that are known to be denser even in the older patients, ie, the posteromedial part of the head. • If anatomically possible, the calcar screw should be used.
• A straight nail design helps to avoid this complication. Instrumentation should allow for precise determination of the nail’s proximal end. • Countersunk proximal screw head reduces the risk of subacromial mechanical impingement. Nail toggling: • With two sizes, the diameter of the nail can be adapted to the diameter of the medullary cavity. • Angular stability, also distal if required, is possible with the angular stable locking system.
230
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
Patient An 88-year-old woman had a fall from standing height. She was living alone and had good cognitive function at baseline. The patient sustained a 3-part greater tuberosity (GT) fracture (HL-G-S), yet the GT was not displaced (Fig 3.1-30a–c). Treatment and outcome The contralateral side was a 2-D reconstruction along the shaft axis; the shape of a straight nail with 8.5 mm in diameter was superimposed (Fig 3.1-30d–f). Note the estimated insertion site at the lateral end of the head fragment, even involving part of the supraspinatus footprint. This situation is critical in the use of antegrade straight nails, as the so-called “fifth anchoring point” does not offer an
a
d
b
CASE 16
Importance of correct entry portal
a dvantage any longer. The authors’ experience has shown that this can be compensated with a slight valgus reduction, or else a different operative procedure, eg, angular stable plate, has to be chosen. The reconstruction showed the entry point problem and consecutive loss of reduction as shown in Fig 3.1-30d–f. Since the insertion was lateral, the “fifth” anchoring point was not established (Fig 3.1-30g–i). Progressive displacement of the reconstruction proved the insufficiency of the implant (Fig 3.1-30j–k). When the implant becomes insufficient, the course of the fracture healing process cannot be predicted (Fig 3.1-30l–m). In this case the patient accepted limited range of motion (90° flexion, external rotation 10°, internal rotation to L1), as she could reach her head and pain was tolerable.
c
e
f
Fig 3.1-30a–m An 88-year-old woman after a fall. a–c X-rays showing a 3-part greater tuberosity (GT) fracture without displaced GT. Note the lateral comminution creating a medium length GT fragment. d–f The contralateral side is a 2-D reconstruction along the shaft axis; the shape of a straight nail with 8.5 mm in diameter is superimposed. Note the estimated insertion site at the lateral end of the head fragment, even involving part of the supraspinatus footprint.
231
Section 3 Fracture management 3.1 Proximal humerus
g
CASE 17
l
h
m
i
k
Fig 3.1-30a–m (cont) An 88-year-old woman after a fall. g–i Due to the later insertion, the “fifth” anchoring point is not established. Note the intraoperatively fractured lesser tuberosity (LT) and the displacement of the LT fragment. j–k X-rays showing progressive displacement of the reconstruction. l–m X-rays showing insufficient implant.
Wrong entry point leads to change of procedure Patient A 72-year-old male patient presented with a low-energy proximal humeral fracture. The patient sustained an unstable, displaced 2-part surgical neck fracture (HGL-S). Comorbidities • Type 2 diabetes • Hypertension • Chronic obstructive pulmonary disease • Osteoporosis • Hyperparathyroidism Treatment and outcome The patient was scheduled for antegrade nailing after informed consent (Fig 3.1-31–b). The computed tomographic scans provided no additional information for the chosen treatment (Fig 3.1-31c–e). The guide wire might have been placed too far anterior or lateral (Fig 3.1-31f–h). If this was the case, the direction of the guide wire was not checked properly. The hollow reamer was then driven in
232
j
without aligning it with the humeral canal, resulting in severe mal alignment when the straight nail was introduced (Fig 3.1-31g). In an effort to achieve better reduction by manipulating the nail, the head fragment cracked and the greater tuberosity split off from the head. Shoulder hemiarthroplasty was then the treatment option selected. Postoperatively, the tuberosities were anatomically fixed (Fig 3.131i–j). In the following weeks, the tuberosities resorbed and the head was migrating cranially (Fig 3.1-31k–l). The patient had only little pain, but the overall result was poor with less than 40° of abduction and flexion, 0° of external rotation and internal rotation to the gluteal region. Conclusion Prosthetic standby may be necessary even in relatively simple cases if severe complications are identified intraoperatively. Prolonged surgical time combined with increased iatrogenic trauma and comorbidities increase the risk of complications like infection, malunion of the tuberosities, or reabsorption.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
a
b
c
d
f
i
e
g
Fig 3.1-31a–l A 72-year-old man with a proximal humeral fracture. a–b X-rays showing unstable, displaced 2-part surgical neck fracture. c–e Computed tomographic scans of the fracture. f–h The guide wire was possibly placed too far anterior or lateral and the hollow reamer then driven in without aligning it with the humeral canal. When introducing the straight nail, severe malalignment must result ( g). i–l P ostoperative anatomical fixation of the tuberosities (i–j), resulting in the tuberosities resorbing and the head migrating cranially in the following weeks (k–l).
h
j
k
l
233
Section 3 Fracture management 3.1 Proximal humerus
4.3.4 Aftercare
Physiotherapy programs following antegrade nailing are similar to the ones following plating. 4.3.5 Outcome
Reports about procedures with the latest generation of nails designs are sparse, especially with regard to the geriatric population. Lopiz et al [63] observed that 9 of 26 patients had rotator cuff symptoms and 11.5% needed reoperations with the MultiLock nail. Hatzidakis et al [66] reported that the fractures of 38 patients aged 65 years on average were treated with an angular stable locked antegrade nail. All fractures healed primarily. The mean follow-up Constant score (and standard deviation) was 71 ± 12 points (range: 37–88 points), with a mean age-adjusted Constant score of 97% (range: 58–119%). All fractures but one healed with a neck-shaft angle of ≥ 125°.
An arthroplasty set should be readily available (Case 17: Fig 3.1-31). The best outcome after fracture arthroplasty results from surgery on day 6 according to the Swedish registry. Thus, fractures that are likely to require a joint replacement procedure should only be operated on if an optimal infrastructure is available. 4.4.2 Approach
The approach most widely used for the anatomical fracture arthroplasty is the deltopectoral approach which can be minimally invasive. When releasing the deltoid muscle close to the clavicle and all the way down to its humeral insertion, the muscle can be retracted far posteriorly in abduction giving full exposure to all aspects of the proximal humerus.
Despite improved locking options and anatomical initial reduction, a minor loss of reduction into varus but without any implant perforations can often be observed. The clinical significance has not yet been investigated sufficiently. Although this may be interpreted as an implant failure, cutthrough or cut-out seem to occur less frequently than in plate fixation. Fracture healing improves the situation, but some malalignment is expected.
As mentioned earlier, performing a tenodesis of the long head of the biceps (LHB) tendon to prevent scarring of an insufficient LHB to the SSC-SSP-interval with subsequent pain and loss of ROM is strongly recommended. Technically, the interval between the SSC and SSP muscles should not be dissected. The LHB can be sutured to the upper pectoralis tendon and cut directly above. The proximal part of the tendon is resected intraarticularly, in case of fracture arthroplasty after the head fragment is removed.
4.4
4.4.3 Typical complications
Shoulder hemiarthroplasty
Because of unpredictable results of shoulder hemiarthroplasty (SHA), surgeons tend to choose between fracture fixation and reverse arthroplasty (Case 18: Fig 3.1-32). If reconstruction is not possible possible, the tuberosities are often not good enough to expect anatomical healing and function with SHA. 4.4.1 Indications
• Shoulder hemiarthroplasty may be indicated in nonreconstructible 3- and 4-part PHFs, head-split injuries and fracture dislocations in older patients. • Suspected AVN does not necessarily require joint replacement as long a stable anatomical reconstruction and fixation can be achieved.
234
• Stable and anatomical reconstruction as well as immediate postoperative physiotherapy are decisive factors in achieving a good result. The decision to go for SHA may be taken intraoperatively if attempts to fix the fracture with preservation of the humeral head fail.
Insufficient fixation, malpositioning and resorption of the tuberosities: • Healing in anatomical position is essential for the result of hemiarthroplasty. “Embracing” fixation of the tuberosities offers superior fixation strength, even more so when applied with flexible cables. • With most systems, the slim metaphyseal part of the stem needs cancellous bone graft underneath the tuberosities to create the bony socket needed for the fixation in the anatomical position. Without this support, the tuberosities are pulled to the metaphyseal stem with the embracing cerclage wires, which results in too low a position. Usually the whole cancellous part of the head fragment is interposed. • Resorption of the tuberosities or consecutive cranialization of the SHA is the reason for inferior functional results comparable to cuff-deficient shoulders. In the worst case scenario an anterosuperior escape with a painful disabling upper extremity might be the result.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Typical course of a shoulder hemiarthroplasty Patient A 77-year-old male patient sustained a 4-part fracture (H-G-L-S) with severely displaced fragments after a fall from a ladder (Fig 3.1-32a–b). Comorbidities • Prostate cancer diagnosed 5 years ago • No relevant other comorbidities present Treatment and outcome The fracture was considered unsuitable for stable anatomical reconstruction and treated with an anatomical head-shaft angle with embracing flexible cables (Fig 3.1-32c–d). The tuberosities were in a good position and the glenohumeral joint was centered in both planes. By
a
e
j
b
f
g
k
that time a lower plexus lesion was diagnosed. Four months later, the proximal humerus was migrating upward indicating an insufficiency of more than two tendons of the rotator cuff (Fig 3.1-32e–f). Heterotopic bone formation was also present indicating the severe soft-tissue trauma. Eighteen months after the fall, the patient had pain; the range of motion of the shoulder was 30° of forward flexion, 0° of external rotation, and when internally rotating he could reach the outer gluteal region (Fig 3.1-32g–i). All parts of the deltoid muscle were working. Almost 3 years after the index trauma the revision to a cemented long stemmed reverse shoulder arthroplasty was carried out, as the patient had been operated in 2010 for a colon carcinoma (Fig 3.1-32j–k). The patient was pain free, function was limited to 80° of forward flexion, 0° of external rotation, and the patient could reach the buttocks.
c
d
h
i
CASE 18
Franz Kralinger, Michael Blauth
Fig 3.1-32a–k A 77-year-old man after a fall from a ladder. a–b X-rays showing a 4-part fracture with severely displaced fragments. c–d The tuberosities were in a good position and the glenohumeral joint centered in both planes. e–f X-rays taken 4 months later: note the ongoing upward migration of the proximal humerus indicating an insufficiency of more than two tendons of the rotator cuff. Heterotopic bone formation is also present indicating the severe soft-tissue trauma. g–i X-rays 18 months after the fall. j–k Almost 3 years after the index trauma, the revision to a cemented long stemmed reverse shoulder arthroplasty was carried out.
235
Section 3 Fracture management 3.1 Proximal humerus
Incorrect height and retroversion of the implant are other factors of major importance. The most accurate reference is the length of the medial head extension, which can be measured accurately. With the shaft dislocated for preparation and implantation of the shaft component, the head can be anatomically reduced and in many cases the anatomical retrotorsion of the given patient can be mimicked perfectly. Too much retrotorsion puts excessive strain on the GT fixation, as it has to be reattached more anteriorly, which potentially leads to more healing problems.
4.4.5 Outcome
4.4.4 Aftercare
Cuff et al [67] followed up 47 patients, aged 74.4 years on average, with 3- and 4-part fractures prospectively for a minimum of 2 years. Three patients (13%) in the hemi arthroplasty group preferred revision to RSA because of failed tuberosity healing and resultant shoulder pseudo paresis. Reverse shoulder arthroplasty resulted in better clinical outcomes and a similar complication rate compared with hemiarthroplasty for the treatment of comminuted PHFs in older adults.
Early passive ROM is essential to avoid stiffness. Immobilization has not been proven to increase the healing rate of the tuberosities but the problem of stiffness.
The outcome of an individual patient is difficult to predict. The results of managing pain are generally good, yet the functional results may vary extremely from very good to unsatisfactory. Anatomical healing of the tuberosities yields good results, outperforming those of RSA (Case 19: Fig 3.1-33). Unfortunately, in contrast to the latter, results of SHA are much less predictable. Radiographic healing can be expected in about 60% of patients and depends greatly on the selection of patients [67].
Ferrel et al [68] performed a systematic review to compare SHA and RSA; they found no significant clinical difference in either the American Shoulder and Elbow Surgeons Shoulder Score (RSA: 64.7, Hemi: 63.0) or the Constant score (RSA: 54.6, Hemi: 58.0). Reverse shoulder arthroplasty was associated with an increased rate of clinical complications (9.6%) and a lower revision rate (0.93%) at short-term to medium-term follow-ups compared with hemiarthroplasty. They concluded that “RSA offers an acceptable surgical option for patients after complex acute PHFs”.
236
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
Patient An 87-year-old man suffered a 3-part greater tuberosity head-split fracture (HL-G-S) resulting from low-energy trauma (Fig 3.1-33a–b). Comorbidities • Condition after cerebral stroke with full recovery and good compliance • Type 2 diabetes Treatment and outcome The supraspinatus tendon insertion was intact, and the muscle belly was without any fatty degeneration or atrophy (Fig 3.1-33c). The reconstruction (Fig 3.1-33d) revealed that the lesser tuberosity was still attached to the head fragment, and due to the pull of the subscapularis (SSC) tendon the shallow head fragment was rotated internally. The muscle belly of the SSC was without atrophy and fatty infiltration.
a
b
c
e
f
g
CASE 19
Healing of tuberosities is unpredictable
The postoperative series showed good alignment of the tuberosities and a centered glenohumeral joint (Fig 3.1-33e–f). At the 4-year follow-up, the healing of the tuberosities was successful (Fig 3.1-33g–h). The joint was centered, the patient was pain free and had an almost full range of motion. The absolute values of the Constant score were 74 points compared to 81 for the healthy uninvolved extremity. Discussion Although the patient’s result is exceptionally favorable, anatomical healing of the tuberosities is unpredictable. Nowadays, a combination of an 87-year-old man, diabetes mellitus and a nonreconstructable head-split fracture would be treated with a reverse shoulder arthroplasty giving a reasonable result independently of the healing of the tuberosities.
d
h
Fig 3.1-33a–h An 87-year-old man with a low-energy trauma. a–b X-rays showing a 3-part greater tuberosity head-split fracture. c–d Supraspinatus tendon insertion was intact, the muscle belly was without any fatty degeneration or atrophy (c). The reconstruction (d) revealed that the lesser tuberosity was still attached to the head fragment, and due to the pull of the subscapularis tendon, the shallow head fragment was rotated internally. e–f The postoperative series showed good alignment of the tuberosities and a centered glenohumeral joint. g–h The 4-year follow-up x-rays showed successful healing of the tuberosities with the joint centered.
237
Section 3 Fracture management 3.1 Proximal humerus
4.5
Reverse shoulder arthroplasty
Due to the unpredictability of SHA results, RSA is used more often in orthogeriatric patients. Especially in patients older than 75 years, there seems to be consensus on the use of a primary RSA if a stable and anatomical reconstruction is not possible. Results are promising and unaffected by the status of the tuberosities healing. Therefore rehabilitation is less critical or even unnecessary. Be aware that a functioning deltoid muscle is essential and in the fracture setting the testing might be difficult at times. Reverse shoulder arthroplasty results in better functional scores than SHA, especially in forward flexion, whereas the RSA result for external rotation is lower than that for SHA. The complication rate seems not considerably higher for patients undergoing RSA compared to SHA, and yet it accounts for lower revision rates. 4.5.1 Indications
In the literature for primary cuff tear arthropathy, arthroplasty with RSA, deltopectoral or superior lateral approaches are described. The superior lateral approach is associated with a lower risk of instability and lower incidence of consecutive acromial and scapular spine fractures. In the acute fracture setting, instability seems to play an inferior role and perfect exposure of the bony glenoid seems essential to lower the risk of malpositioning of the metaglene and consecutive loosening or scapular notching. Therefore the deltopectoral approach remains the main approach in acute fracture care, as it leaves all options open, including complex revision if needed. 4.5.4 Implant selection
Currently, there are more than 20 different systems on the market. Medialized or lateralized RSA systems are available. As all systems promote their given designs, it is difficult to choose.
• Complex 3- and 4-part fractures that are not suitable for a stable and anatomical reconstruction due to poor bone quality. • Fractures that have “short” and comminuted tuberosities and a high risk of losing the tuberosities when treated with SHA. • Head-split fractures are selected for SHA or RSA under the same criteria as already mentioned. • Preexisting computed tomographic angiography (Case 20: Fig 3.1-34).
4.5.5 Fixation of the tuberosities
4.5.2 Contraindications
For detailed description of the technique, see Reuther et al [69].
• Substantial glenoid bone loss or lack of glenoid bone stock might be a contraindication. Complex bony reconstruction and revision of metaglene fixation is not yet being discussed for acute fracture care. • Axillary nerve deficiency or palsy. An intact deltoid muscle is vital for a functional RSA, therefore axillary nerve investigation is crucial in the acute fracture patient. Reverse shoulder arthroplasty is contraindicated in this situation as the increased risk for instability and the expected low functional result do not outweigh the risk of the intervention.
238
4.5.3 Approach
Rival concepts concerning repair of the tuberosities fuel an ongoing debate. Total resection of the tuberosities to full repair maintaining the full cuff (including SSP) are described. In order to gain more external rotation, some surgeons advocate cutting the SSC tendon or resecting the LT and peforming an isolated repair of the GT. The authors resect the SSP and the upper part of ISP with their bony attachment and fix the LT and the remaining GT with cerclage sutures to the prosthesis.
4.5.6 Typical complications
• Scapular notching is the most common complication, responsible for polyethylene wear and consequent loosening. • Acromial or scapular spine insufficiency fractures due to the high stress levels on the implant-bone-interface are reported. • The risk of infection is supposed to be higher in RSA fracture arthroplasty compared to SHA; however this could not be proven in recent systematic reviews. • Neuropathy and instability are other typical complications.
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas
Franz Kralinger, Michael Blauth
Three recent systematic reviews and metaanalyses [68–71] compare RSA and SHA for acute fracture care. The results concerning complication rates, functional outcome, revision rates, and cost factors are conflicting. According to Namdari et al [71] both systems have the potential to restore pain-free function. They believe that the higher complication rates and costs in RSA should be considered individually. Mata-Fink et al [70] saw better forward
Patient A 78-year-old woman fell from a standing height, sustaining a 2-part surgical neck fracture (HGL-S) (Fig 3.1-34a–c). One year prior to the fall the patient was treated with a proximal femoral nail antirotation for a pertrochanteric fracture. No relevant comorbidities are present and the patient is not cognitively impaired. Treatment and outcome The patient sustained a 2-part surgical neck fracture (HGL-S) (Fig 3.1-34a–c). She was known from the shoulder outpatient department (see Fig 3.1-2, Fig 3.1-3, Fig 3.1-4). Reconstruction in the presence of a preexisting computed tomographic angiography in a 78-year-old patient was not indicated (Fig 3.1-34d–f).
a
b
flexion and functional outcome of RSA, whereas in their study complication rates did not differ substantially. Ferrel et al [68] stated better forward flexion, lower external rotation, but no statistical differences according to the American Shoulder and Elbow Surgeons Shoulder Score (RSA: 64.7, Hemi: 63.0) or the Constant score (RSA: 54.6, Hemi: 58.0). Complication rates for RSA were higher, yet in contrast with this finding revision rates were lower compared with the SHA.
The fracture was addressed with a reverse shoulder arthroplasty and the tuberosities were reattached in an overlapping mode (tile technique) with embracing sutures. The supraspinatus muscle and upper part of infraspintus muscle inserted on top of the greater tuberosity was released (Fig 3.1-34g–h).
CASE 20
4.5.7 Outcome
At the 2-year follow-up the patient herself was satisfied, as her range of motion was 140° of flexion, external rotation in 0° of abduction was -10°, and in internal rotation she could reach the lower lumbar spine. She continued to live independently and could even do some gardening (Fig 3.1-34i–j).
c
Fig 3.1-34a–j A 78-year-old woman after a fall from standing height. a–c X-rays showing a 2-part surgical neck fracture.
239
Section 3 Fracture management 3.1 Proximal humerus
d
e
g
h
f
i
j
Fig 3.1-34a–j (cont) A 78-year-old woman after a fall from standing height. d–f A preexisting computed tomographic angiography. g–h Tuberosities reattached in an overlapping mode with embracing sutures. The supraspinatus muscle and upper part of infraspintus muscle inserted on top of the greater tuberosity is released. i–j Two-year follow-up x-rays.
4.6
Summary
Reconstruction of complex orthogeriatric PHF remains challenging and high complication rates are reported. In selected patients and with stable and anatomical reconstruction good results can be achieved. Supportive techniques like allografts or implant augmentation with cement may improve the outcome. Anatomical SHA is associated with inconsistent results depending on the status of the tuberosities. In more recent studies with modern stem designs and adequate embracing fixation of the tuberosities, results are comparable with RSA, consistently better regarding external rotation. Reverse shoulder arthroplasty is increasing in use for primary fracture care. Many surgeons advocate the exclusive use of this type of joint replacement in patients aged 70 years and over, when treating complex 3- and 4-part fractures. Outcome data are conflicting and the need for prospective randomized studies is obvious.
240
Osteoporotic Fracture Care Michael Blauth, Stephen L Kates, Joseph A Nicholas